技术领域Technical Field
本公开涉及图像处理技术领域，特别地涉及一种病灶分割模型的训练方法、装置、设备及存储介质。The present disclosure relates to the field of image processing technology, and in particular to a training method, device, equipment and storage medium for a lesion segmentation model.
背景技术Background Art
随着人工智能（Artificial Intelligence，AI）技术的飞速发展，AI在病灶自动检测方面发挥着越来越重要的作用，大大减轻了医生的负担。但是，病灶位置的多样性和形态的复杂性给AI模型的训练带来了挑战。要想取得令人满意的效果，需要大规模数据的支持。然而，获取足够数量且质量较好的数据是一项艰巨的任务，这使得病灶分割模型训练过程变得更加困难和耗时，如何实现提高模型的病灶检出率的模型训练至关重要。With the rapid development of artificial intelligence (AI) technology, AI plays an increasingly important role in automatic lesion detection, greatly reducing the burden on doctors. However, the diversity of lesion locations and the complexity of morphology pose challenges to the training of AI models. To achieve satisfactory results, large-scale data support is required. However, obtaining sufficient and high-quality data is a difficult task, which makes the lesion segmentation model training process more difficult and time-consuming. How to achieve model training to improve the lesion detection rate of the model is crucial.
发明内容Summary of the invention
本公开提供一种病灶分割模型的训练方法、装置、设备及存储介质，以提高The present invention provides a training method, device, equipment and storage medium for a lesion segmentation model to improve
模型在病灶分割中的检出率。Detection rate of the model in lesion segmentation.
第一方面，本公开提供了一种病灶分割模型的训练方法，包括：In a first aspect, the present disclosure provides a method for training a lesion segmentation model, comprising:
读取和预处理数据集，所述数据集包含CT数据，所述CT数据包括CT图像及对应的标注图像；Reading and preprocessing a data set, wherein the data set includes CT data, and the CT data includes a CT image and a corresponding annotated image;
基于读取和预处理的结果、第一损失函数和第二损失函数，训练第一病灶分割模型和第二病灶分割模型；Based on the results of reading and preprocessing, the first loss function and the second loss function, training a first lesion segmentation model and a second lesion segmentation model;
其中，所述第一病灶分割模型和所述第二病灶分割模型均为神经网络模型，所述第一病灶分割模型的训练使用第一损失函数，所述第二病灶分割模型的前预设轮训练使用第一损失函数，所述第二病灶分割模型的其他轮训练使用第二损失函数，以使训练得到的第二病灶分割模型分割的病灶尺寸小于训练得到的第一病灶分割模型分割的病灶尺寸。Among them, the first lesion segmentation model and the second lesion segmentation model are both neural network models, the first lesion segmentation model is trained using a first loss function, the first preset round of training of the second lesion segmentation model uses the first loss function, and other rounds of training of the second lesion segmentation model use a second loss function, so that the lesion size segmented by the trained second lesion segmentation model is smaller than the lesion size segmented by the trained first lesion segmentation model.
在一些实施例中，所述第二损失函数表达式如下：In some embodiments, the second loss function is expressed as follows:
其中，表示第i个病灶的损失权重，的数值基于第i个病灶占有的像素数量确定，第i个病灶占有的像素数量越少，的数值越大；表示第i个病灶的交叉熵损失；表示第i个病灶的Tversky损失；a0和b0分别表示和的系数；表示未正确检测的病灶损失，未正确检测的病灶越多，的数值越大。in, represents the loss weight of thei- th lesion, The value of is determined based on the number of pixels occupied by the i- th lesion. The smaller the number of pixels occupied by the i- th lesion, The larger the value of; represents the cross entropy loss of thei- th lesion; represents the Tversky loss of theith lesion;a 0 andb 0 represent and The coefficient of Indicates the loss of incorrectly detected lesions. The more incorrectly detected lesions are, The larger the value.
在一些实施例中，通过如下计算式确定第i个病灶的损失权重：In some embodiments, the loss weight of theith lesion is determined by the following calculation formula:
其中，a1和b1均表示可调节的参数，通过调节a1和b1能够控制第i个病灶的损失权重；表示第i个病灶占有的像素数量，表示所有病灶占有的像素数量。Among them,a 1 andb 1 are adjustable parameters. By adjustinga 1 andb 1, the loss weight of thei- th lesion can be controlled; represents the number of pixels occupied by thei -th lesion, Indicates the number of pixels occupied by all lesions.
在一些实施例中，通过如下计算式确定未正确检测的病灶损失：In some embodiments, the incorrectly detected lesion loss is determined by the following calculation:
其中，a和b均表示可调节的参数，通过调节a和b能够控制未正确检测的病灶的惩罚大小；numneg表示未正确检测的病灶数量。Wherein,a andb are adjustable parameters, and the penalty for incorrectly detected lesions can be controlled by adjustinga andb ;numneg represents the number of incorrectly detected lesions.
在一些实施例中，所述未正确检测的病灶数量按照如下方法得到：在所述第二病灶分割模型每一次训练结束时，计算数据集的标注图像中每一个病灶和预测标注图像中每一个病灶的Dice指标值，基于计算得到的Dice指标值统计得到未正确检测的病灶数量。In some embodiments, the number of incorrectly detected lesions is obtained according to the following method: at the end of each training of the second lesion segmentation model, the Dice index value of each lesion in the annotated image of the data set and each lesion in the predicted annotated image is calculated, and the number of incorrectly detected lesions is obtained based on the calculated Dice index value.
在一些实施例中，所述未正确检测的病灶数量按照如下方法得到：In some embodiments, the number of incorrectly detected lesions is obtained as follows:
在所述第二病灶分割模型每一次训练结束时，从数据集的标注图像中逐个抽取病灶，将抽取的病灶分别与预测标注图像中的病灶进行Dice指标值计算，将计算得到的Dice指标值记录到第一列表中，第一列表的行表示数据集的标注图像中的病灶，第一列表的列表示预测标注图像中的病灶；At the end of each training of the second lesion segmentation model, lesions are extracted one by one from the annotated images of the data set, Dice index values are calculated for the extracted lesions and the lesions in the predicted annotated images, and the calculated Dice index values are recorded in a first list, where the rows of the first list represent the lesions in the annotated images of the data set, and the columns of the first list represent the lesions in the predicted annotated images;
根据Dice指标值的大小对第一列表的每一列分别进行排序；Sort each column of the first list according to the value of the Dice index;
针对第一列表的每一列确定最大的Dice指标值是否不小于第一数值；Determine for each column of the first list whether the maximum Dice index value is not less than the first value;
若是，则该Dice指标值对应的病灶被所述第二病灶分割模型正确检测，所述未正确检测的病灶数量的值不变；If yes, the lesion corresponding to the Dice index value is correctly detected by the second lesion segmentation model, and the value of the number of lesions that are not correctly detected remains unchanged;
否则，该病灶未被所述第二病灶分割模型正确检测，所述未正确检测的病灶数量的值加一；Otherwise, the lesion is not correctly detected by the second lesion segmentation model, and the value of the number of lesions that are not correctly detected is increased by one;
统计得到最终的未正确检测的病灶数量的值。The final value of the number of incorrectly detected lesions is obtained by counting.
在一些实施例中，所述读取和预处理数据集，包括：In some embodiments, the reading and preprocessing of the data set comprises:
打乱所述数据集中各CT数据的顺序并读取打乱顺序后的各CT数据；Disrupting the order of each CT data in the data set and reading each CT data after the order is disrupted;
将读取的各CT数据中的CT图像对应标注图像中的病灶标注进行膨胀运算；Performing dilation operation on the CT image in each CT data read corresponding to the lesion annotation in the annotation image;
将每一例CT图像及其标注图像的方向转换为RAI方向；Convert the orientation of each CT image and its annotated image into RAI orientation;
对每一例CT图像，进行窗宽窗位的调整；For each CT image, adjust the window width and window position;
对每一例CT图像进行像素归一化处理；Perform pixel normalization on each CT image;
将每一例CT图像及对应标注图像，根据感兴趣区域进行三维裁剪；Each CT image and the corresponding annotated image are three-dimensionally cropped according to the region of interest;
将裁剪后的CT图像及对应标注图像的X轴和Y轴进行尺度缩放；Scale the X-axis and Y-axis of the cropped CT image and the corresponding annotated image;
将每一例CT图像及对应标注图像进行切片处理，得到一组切片图像；Each CT image and the corresponding annotated image are sliced to obtain a set of slice images;
将各CT数据对应的所有组切片图像打乱顺序，得到读取和预处理的结果。All the groups of slice images corresponding to each CT data are disrupted to obtain the reading and preprocessing results.
在一些实施例中，所述基于读取和预处理的结果、第一损失函数和第二损失函数，训练第一病灶分割模型和第二病灶分割模型，训练过程如下：In some embodiments, the first lesion segmentation model and the second lesion segmentation model are trained based on the reading and preprocessing results, the first loss function and the second loss function, and the training process is as follows:
从读取和预处理的结果中按批次抽取图像数据进行在线数据增强处理；Extract image data in batches from the results of reading and preprocessing to perform online data augmentation processing;
基于在线数据增强后的图像数据进行第一病灶分割模型或第二病灶分割模型的训练，并计算当前轮中每一次的损失函数值；Training the first lesion segmentation model or the second lesion segmentation model based on the image data after online data enhancement, and calculating the loss function value each time in the current round;
计算当前平均损失函数值；Calculate the current average loss function value;
计算当前轮中每一次的损失函数值与当前平均损失函数值的大小：若当前轮中每一次的损失函数值≥当前平均损失函数值，则将当前批次图像数据存储到第二列表中；Calculate the magnitude of the loss function value of each time in the current round and the current average loss function value: if the loss function value of each time in the current round is ≥ the current average loss function value, store the current batch of image data in the second list;
当前轮图像数据训练结束后，读取第二列表中图像数据进行补充训练，若第二列表中任一批次图像数据对应的损失函数值≥当前平均损失函数值，则将该批次图像数据存储到第三列表中；After the current round of image data training is completed, the image data in the second list is read for supplementary training. If the loss function value corresponding to any batch of image data in the second list is ≥ the current average loss function value, the batch of image data is stored in the third list;
每一轮训练结束后，将第二列表中的内容更新为第三列表中的内容；After each round of training, the content in the second list is updated to the content in the third list;
判断当前平均损失函数值是否大于预设阈值：若是，则继续进行下一轮训练；否则结束训练，保存平均损失函数值最小时的第一病灶分割模型或第二病灶分割模型。Determine whether the current average loss function value is greater than a preset threshold: if so, continue with the next round of training; otherwise, end the training and save the first lesion segmentation model or the second lesion segmentation model when the average loss function value is the smallest.
在一些实施例中，所述读取和预处理数据集之前，还包括：对数据集中的CT数据进行离线数据增强，以对数据集的CT数据扩增。In some embodiments, before reading and preprocessing the data set, the method further includes: performing offline data enhancement on the CT data in the data set to amplify the CT data in the data set.
第二方面，本公开提供了一种病灶分割方法，包括：In a second aspect, the present disclosure provides a lesion segmentation method, comprising:
获取CT图像；Acquire CT images;
将CT图像进行切片，得到一组切片图像；Slice the CT image to obtain a set of slice images;
将该组切片图像依次放入第一病灶分割模型和第二病灶分割模型中进行计算，得到第一病灶分割模型分割结果和第二病灶分割模型分割结果，所述第一病灶分割模型和所述第二病灶分割模型是基于第一方面所述病灶分割模型的训练方法训练得到的。This group of slice images is sequentially placed into the first lesion segmentation model and the second lesion segmentation model for calculation to obtain the segmentation results of the first lesion segmentation model and the second lesion segmentation model, wherein the first lesion segmentation model and the second lesion segmentation model are trained based on the training method of the lesion segmentation model described in the first aspect.
在一些实施例中，所述的病灶分割方法还包括：对第一病灶分割模型分割结果和第二病灶分割模型分割结果求并集或者交集。In some embodiments, the lesion segmentation method further includes: obtaining a union or intersection of a segmentation result of the first lesion segmentation model and a segmentation result of the second lesion segmentation model.
第三方面，本公开提供了一种病灶分割模型的训练装置，包括：In a third aspect, the present disclosure provides a training device for a lesion segmentation model, comprising:
数据处理模块，用于读取和预处理数据集，所述数据集包含CT数据，CT数据包括CT图像及对应的标注图像；A data processing module, used for reading and preprocessing a data set, wherein the data set includes CT data, and the CT data includes a CT image and a corresponding annotated image;
模型训练模块，用于基于读取和预处理的结果、第一损失函数和第二损失函数，训练第一病灶分割模型和第二病灶分割模型；其中，所述第一病灶分割模型和所述第二病灶分割模型均为神经网络模型，所述第一病灶分割模型的训练使用第一损失函数，所述第二病灶分割模型的前预设轮训练使用第一损失函数，所述第二病灶分割模型的其他轮训练使用第二损失函数，以使训练得到的第二病灶分割模型分割的病灶尺寸小于训练得到的第一病灶分割模型分割的病灶尺寸。A model training module is used to train a first lesion segmentation model and a second lesion segmentation model based on the results of reading and preprocessing, a first loss function and a second loss function; wherein, the first lesion segmentation model and the second lesion segmentation model are both neural network models, the first lesion segmentation model is trained using a first loss function, the first preset round of training of the second lesion segmentation model uses the first loss function, and other rounds of training of the second lesion segmentation model use a second loss function, so that the size of the lesion segmented by the trained second lesion segmentation model is smaller than the size of the lesion segmented by the trained first lesion segmentation model.
第四方面，本公开提供了一种病灶分割装置，包括：In a fourth aspect, the present disclosure provides a lesion segmentation device, comprising:
图像获取模块，用于获取CT图像；An image acquisition module, used for acquiring CT images;
图像切片模块，用于将CT图像进行切片，得到一组切片图像；An image slicing module is used to slice the CT image to obtain a set of slice images;
病灶分割模块，用于将该组切片图像依次放入第一病灶分割模型和第二病灶分割模型中进行计算，得到第一病灶分割模型分割结果和第二病灶分割模型分割结果，所述第一病灶分割模型和所述第二病灶分割模型是基于第三方面所述病灶分割模型的训练装置训练得到的。A lesion segmentation module is used to sequentially place the group of slice images into a first lesion segmentation model and a second lesion segmentation model for calculation to obtain a segmentation result of the first lesion segmentation model and a segmentation result of the second lesion segmentation model, wherein the first lesion segmentation model and the second lesion segmentation model are trained by a training device for the lesion segmentation model described in the third aspect.
第五方面，本公开提供了一种计算机设备，包括存储器、处理器及存储在存储器上的计算机程序，所述处理器执行所述计算机程序以实现第一方面或第二方面所述方法的步骤。In a fifth aspect, the present disclosure provides a computer device, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method described in the first aspect or the second aspect.
第六方面，本公开提供了一种计算机可读存储介质，其上存储有计算机程序，该计算机程序被处理器执行时实现第一方面或第二方面所述方法的步骤。In a sixth aspect, the present disclosure provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect or the second aspect.
第七方面，本公开提供了一种计算机程序产品，包括计算机程序/指令，该计算机程序被处理器执行时实现第一方面或第二方面所述方法的步骤。In a seventh aspect, the present disclosure provides a computer program product, comprising a computer program/instructions, which, when executed by a processor, implements the steps of the method described in the first aspect or the second aspect.
本公开提供的一种病灶分割模型的训练方法、装置、设备及存储介质，通过使用第一损失函数训练第一病灶分割模型，使用第一损失函数训练第二病灶分割模型的前预设轮，使用第二损失函数训练第二病灶分割模型的其他轮，使第二病灶分割模型更加关注和适用于小尺寸病灶的分割，通过不同训练方式训练两种神经网络模型，分别用来解决脏器中不同尺寸病灶的分割问题，显著提高了小尺寸病灶的检出率。The present disclosure provides a training method, apparatus, device and storage medium for a lesion segmentation model. The first lesion segmentation model is trained by using a first loss function, the first preset round of the second lesion segmentation model is trained by using the first loss function, and the other rounds of the second lesion segmentation model are trained by using a second loss function, so that the second lesion segmentation model pays more attention to and is more suitable for the segmentation of small-sized lesions. Two neural network models are trained by different training methods to solve the segmentation problem of lesions of different sizes in organs, respectively, thereby significantly improving the detection rate of small-sized lesions.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
在下文中将基于实施例并参考附图来对本公开进行更详细的描述：The present disclosure will be described in more detail below based on embodiments and with reference to the accompanying drawings:
图1为本公开实施例提供的一种病灶分割模型的训练方法的流程示意图。FIG1 is a flow chart of a training method for a lesion segmentation model provided in an embodiment of the present disclosure.
图2为本公开实施例提供的第i个病灶的损失权重的曲线示意图。FIG2 is a schematic diagram of a curve showing the loss weight of theith lesion provided in an embodiment of the present disclosure.
图3为本公开实施例提供的未正确检测的病灶数量的损失值的曲线示意图。FIG3 is a schematic diagram of a curve showing the loss value of the number of incorrectly detected lesions provided in an embodiment of the present disclosure.
图4为本公开实施例提供的一种病灶分割方法的流程示意图。FIG4 is a flow chart of a lesion segmentation method provided in an embodiment of the present disclosure.
在附图中，相同的部件使用相同的附图标记，附图并未按照实际的比例绘制。In the drawings, the same reference numerals are used for the same components, and the drawings are not drawn to scale.
具体实施方式DETAILED DESCRIPTION
为了使本技术领域的人员更好地理解本公开的技术方案，并对本公开如何应用技术手段来解决技术问题，并达到相应技术效果的实现过程能充分理解并据以实施，下面将结合本公开实施例中的附图，对本公开的实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本公开一部分的实施例，而不是全部的实施例。本公开的实施例以及实施例中的各个特征，在不相冲突前提下可以相互结合，所形成的技术方案均在本公开的保护范围之内。基于本公开中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都应当属于本公开保护的范围。In order to enable those skilled in the art to better understand the technical solution of the present disclosure, and to fully understand and implement how the present disclosure applies technical means to solve technical problems and achieve the corresponding technical effects, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only embodiments of a part of the present disclosure, not all of the embodiments. The embodiments of the present disclosure and the various features in the embodiments can be combined with each other without conflict, and the technical solutions formed are all within the scope of protection of the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of the present disclosure.
需要说明的是，本公开的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便这里描述的本公开的实施例能够以除了在这里图示或描述的那些以外的顺序实施。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.
需要说明的是，在附图的流程图示出的步骤可以在诸如一组计算机可执行指令的计算机系统中执行，并且，虽然在流程图中示出了逻辑顺序，但是在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤。It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.
实施例一Embodiment 1
图1为本公开实施例提供的一种病灶分割模型的训练方法的流程示意图。如图1所示，一种病灶分割模型的训练方法，包括：FIG1 is a flow chart of a training method for a lesion segmentation model provided by an embodiment of the present disclosure. As shown in FIG1 , a training method for a lesion segmentation model includes:
步骤S11、读取和预处理数据集，数据集包含CT数据，CT数据包括CT图像及对应的标注图像。Step S11 : reading and preprocessing a data set, where the data set includes CT data, and the CT data includes a CT image and a corresponding annotated image.
步骤S12、基于读取和预处理的结果、第一损失函数和第二损失函数，训练第一病灶分割模型和第二病灶分割模型。Step S12: training the first lesion segmentation model and the second lesion segmentation model based on the reading and preprocessing results, the first loss function and the second loss function.
其中，第一病灶分割模型和第二病灶分割模型均为神经网络模型，第一病灶分割模型的训练使用第一损失函数，第二病灶分割模型的前预设轮训练使用第一损失函数，第二病灶分割模型的其他轮训练使用第二损失函数，以使训练得到的第二病灶分割模型分割的病灶尺寸小于训练得到的第一病灶分割模型分割的病灶尺寸。Among them, the first lesion segmentation model and the second lesion segmentation model are both neural network models, the first lesion segmentation model is trained using a first loss function, the first preset round of training of the second lesion segmentation model uses the first loss function, and other rounds of training of the second lesion segmentation model use the second loss function, so that the lesion size segmented by the trained second lesion segmentation model is smaller than the lesion size segmented by the trained first lesion segmentation model.
本实施例中，通过使用第一损失函数训练第一病灶分割模型，使用第一损失函数训练第二病灶分割模型的前预设轮，使用第二损失函数训练第二病灶分割模型的其他轮，使第二病灶分割模型更加关注和适用于小尺寸病灶的分割，通过不同训练方式训练两种神经网络模型，分别用来解决脏器中不同尺寸病灶的分割问题，显著提高了小尺寸病灶的检出率。In this embodiment, the first lesion segmentation model is trained using the first loss function, the first preset rounds of the second lesion segmentation model are trained using the first loss function, and the other rounds of the second lesion segmentation model are trained using the second loss function, so that the second lesion segmentation model pays more attention to and is more suitable for the segmentation of small-sized lesions. Two neural network models are trained by different training methods to solve the segmentation problems of lesions of different sizes in organs, respectively, thereby significantly improving the detection rate of small-sized lesions.
在读取和预处理数据集之前，本实施例需要搭建模型训练环境，例如：Before reading and preprocessing the data set, this embodiment needs to build a model training environment, for example:
选择公开号为CN117058158A，名称为一种轻量级医学图像分割方法、装置、介质及电子设备的专利申请中的轻量级医学图像分割模型作为待训练的神经网络模型，标记为FDDW_ConvLSTM_SuperLight_3D，这一待训练的神经网络模型在本申请中分别作为待训练的第一病灶分割模型和第二病灶分割模型。A lightweight medical image segmentation model in a patent application with publication number CN117058158A and titled A lightweight medical image segmentation method, device, medium and electronic device is selected as the neural network model to be trained, marked as FDDW_ConvLSTM_SuperLight_3D. This neural network model to be trained is used as the first lesion segmentation model and the second lesion segmentation model to be trained in this application.
选择AdamW作为优化器，将其学习率定义为：AdamW is selected as the optimizer and its learning rate is defined as:
~， ~ ,
权重衰减定义为：Weight decay is defined as:
~。 ~ .
其中，batch_size为一次训练的批数大小。Among them, batch_size is the batch size of one training.
应当理解的是，在实际应用本实施例的方法时，可以选择任意神经网络模型进行训练，并不限于上述专利申请中的模型。It should be understood that when the method of this embodiment is actually applied, any neural network model can be selected for training and is not limited to the model in the above-mentioned patent application.
以前述专利申请中的轻量级医学图像分割模型为例，采用本实施例的方法进行训练后，分别对肺结节和肝脏病灶进行了分割实验，实验结果如下表所示：Taking the lightweight medical image segmentation model in the aforementioned patent application as an example, after training using the method of this embodiment, segmentation experiments were conducted on lung nodules and liver lesions, respectively. The experimental results are shown in the following table:
表1Table 1
表2Table 2
可以看出，采用本实施例的方法训练得到的第二病灶分割模型，对于小病灶或微小病灶的检出率有了明显提升。其中，对于肺结节来说，直径在3mm~8mm之间属于小病灶，对于肝脏病灶来说，直径小于20mm属于微小病灶。It can be seen that the second lesion segmentation model trained by the method of this embodiment has significantly improved the detection rate of small lesions or micro lesions. For lung nodules, lesions with a diameter between 3 mm and 8 mm are small lesions, and for liver lesions, lesions with a diameter less than 20 mm are micro lesions.
在训练第一病灶分割模型时，只需要使用第一损失函数。When training the first lesion segmentation model, only the first loss function needs to be used .
在训练第二病灶分割模型时，前I轮训练使用第一损失函数，后J轮训练使用第二损失函数，第二病灶分割模型的总训练轮数是I+J。在实际应用中，I的取值范围可以是1~20，一种优选情况下，I=10；J的取值范围可以是1~100，一种优选情况下，J=50。When training the second lesion segmentation model, the first round of training uses the first loss function , the second loss function is used for the next J rounds of training , the total number of training rounds of the second lesion segmentation model is I + J. In practical applications, the value range of I can be 1-20, and in a preferred case, I=10; the value range of J can be 1-100, and in a preferred case, J=50.
第一病灶分割模型和第二病灶分割模型前I轮的第一损失函数定义如下：The first loss function of the first lesion segmentation model and the first I rounds of the second lesion segmentation model The definition is as follows:
其中，表示交叉熵损失函数，表示Tversky损失函数；m和n分别表示和的系数，m和n的范围是0.0~1.0，考虑到病灶本身的尺寸较小，在保证模型稳定训练的同时，也需要保证病灶的检出率，设置m为0.7，保证模型稳定训练，设置n为0.5，以达到提高病灶检出率的要求。in, represents the cross entropy loss function, represents the Tversky loss function; m and n represent and The coefficients of m and n range from 0.0 to 1.0. Considering the small size of the lesion itself, it is necessary to ensure the detection rate of the lesion while ensuring the stable training of the model. m is set to 0.7 to ensure the stable training of the model, and n is set to 0.5 to achieve the requirement of improving the lesion detection rate.
的表达式为： The expression is:
其中，n表示样本总数，x表示样本，集合表示医生绘制的器官病灶标注数据，集合表示模型生成的器官病灶标注数据。Among them, n represents the total number of samples, x represents the sample, and the set Represents the organ lesion annotation data drawn by doctors, set Represents the organ lesion annotation data generated by the model.
的表达式为： The expression is:
其中，集合表示模型生成的器官病灶标注数据，集合表示医生绘制的器官病灶标注数据；表示假阳性FP，表示假阴性FN，表示平滑系数，优选地，取值为时计算较为准确。Among them, the collection Represents the organ lesion annotation data generated by the model, set Represents the organ lesion annotation data drawn by doctors; represents a false positive FP, Indicates false negative FN, represents the smoothing coefficient, preferably, the value is The calculation is more accurate.
Tversky损失函数具有两个参数α和β，其优势是通过调整α和β两个参数，可以控制假阳性FP和假阴性FN之间的平衡，进而影响小分割区域和大分割区域的分割准确度。The Tversky loss function has two parameters α and β. Its advantage is that by adjusting the two parameters α and β, the balance between false positive FP and false negative FN can be controlled, thereby affecting the segmentation accuracy of small segmentation areas and large segmentation areas.
参数α取值范围是0.2~0.35，β的取值范围是0.6~0.75，在一个示例中，α=0.3，β=0.7。参数α和参数β的设置增加了图像分割灵敏度的权重，有利于肺结节等较小病灶区域的分割。The parameter α ranges from 0.2 to 0.35, and the parameter β ranges from 0.6 to 0.75. In one example, α = 0.3 and β = 0.7. The setting of the parameters α and β increases the weight of the image segmentation sensitivity, which is beneficial to the segmentation of smaller lesion areas such as lung nodules.
在一些实施方式中，第二损失函数表达式如下：In some implementations, the second loss function is expressed as follows:
其中，表示第i个病灶的损失权重，的数值基于第i个病灶占有的像素数量确定，第i个病灶占有的像素数量越少，的数值越大；表示第i个病灶的交叉熵损失；表示第i个病灶的Tversky损失；a0和b0分别表示和的系数；表示未正确检测的病灶损失，未正确检测的病灶越多，的数值越大。a0和b0的取值范围是0.0~1.0，考虑到病灶本身尺寸较小，在保证模型稳定训练的同时，也需要保证病灶的检出率，设置a0为0.7，保证模型稳定训练，设置b0为0.5，以达到提高病灶检出率的要求。in, represents the loss weight of thei- th lesion, The value of is determined based on the number of pixels occupied by the i- th lesion. The smaller the number of pixels occupied by the i- th lesion, The larger the value of; represents the cross entropy loss of thei- th lesion; represents the Tversky loss of theith lesion;a 0 andb 0 represent and The coefficient of Indicates the loss of incorrectly detected lesions. The more incorrectly detected lesions are, The larger the value ofa0 andb0 is, the range of a0 and b0 is 0.0~1.0. Considering the small size of the lesion itself, it is necessary to ensure the detection rate of the lesion while ensuring the stable training of the model.a0 is set to 0.7 to ensure the stable training of the model, andb0 is set to 0.5 to achieve the requirement of improving the detection rate of the lesion.
在一些实施例中，通过如下计算式确定第i个病灶的损失权重：In some embodiments, the loss weight of theith lesion is determined by the following calculation formula:
其中，a1和b1均表示可调节的参数，通过调节a1和b1能够控制第i个病灶的损失权重；表示第i个病灶占有的像素数量，表示所有病灶占有的像素数量。a1的取值范围是2~20，b1的取值范围是2~10。优选的，当a1=10,b1=3时，效果较好。Among them,a 1 andb 1 are adjustable parameters. By adjustinga 1 andb 1, the loss weight of thei- th lesion can be controlled; represents the number of pixels occupied by thei -th lesion, Indicates the number of pixels occupied by all lesions. The value range ofa1 is 2~20, and the value range ofb1 is 2~10. Preferably, whena1 =10 andb1 =3, the effect is better.
为了适应不同器官的病灶分割任务，这里给出上述第i个病灶的损失权重计算式中a1和b1参数的设置方法的一个示例：In order to adapt to the lesion segmentation tasks of different organs, an example of setting the parametersa1 andb1 in the above loss weight calculation formula for thei -th lesion is given here:
设置参数时，固定a1=10不变，调节b1的大小。当a1固定为10，而b1分别为2或5时，该函数的曲线如图2所示。When setting parameters, fixa 1=10 and adjust the size ofb 1. Whena 1 is fixed to 10 andb 1 is 2 or 5, the function The curve is shown in Figure 2.
其中，曲线1参数为a1=10和b1=2，曲线2参数为a1=10和b1=5。从图2中可以看出，b1越大，曲线越陡峭，体积较小的病灶的loss权重越大。b1越小，曲线越平缓，体积较小的病灶的loss权重越小。Among them, the parameters of curve 1 area 1=10 andb 1=2, and the parameters of curve 2 area 1=10 andb 1=5. As can be seen from Figure 2, the largerthe b 1, the steeper the curve, and the larger the loss weight of the smaller lesion. The smallerthe b 1, the flatter the curve, and the smaller the loss weight of the smaller lesion.
在一些实施方式中，通过如下计算式确定未正确检测的病灶损失：In some embodiments, incorrectly detected lesion loss is determined by the following calculation:
其中，a和b均表示可调节的参数，通过调节a和b能够控制未正确检测的病灶的惩罚大小；numneg表示未正确检测的病灶数量。通过在第二损失函数中采用第i个病灶的损失权重和未正确检测的病灶损失两种控制手段来调节小尺寸病灶的惩罚程度，以控制小尺寸病灶的检出率。a的取值范围是2~20，b的取值范围是0.1~10.0。优选的，当a=10，b=0.5时，效果较好。Where,a andb are adjustable parameters. By adjustinga andb , the penalty for incorrectly detected lesions can be controlled.numneg represents the number of incorrectly detected lesions. By using the loss weight of theith lesion in the second loss function and incorrectly detected lesion loss Two control means are used to adjust the degree of penalty for small-sized lesions to control the detection rate of small-sized lesions. The value range ofa is 2-20, and the value range ofb is 0.1-10.0. Preferably, whena = 10 andb = 0.5, the effect is better.
在一些实施方式中，未正确检测的病灶数量可以按照如下方法得到：在第二病灶分割模型每一次训练结束时，计算数据集的标注图像中每一个病灶和预测标注图像中每一个病灶的Dice指标值，基于计算得到的Dice指标值统计得到未正确检测的病灶数量。In some embodiments, the number of incorrectly detected lesions can be obtained as follows: at the end of each training of the second lesion segmentation model, the Dice index value of each lesion in the annotated image of the data set and each lesion in the predicted annotated image is calculated, and the number of incorrectly detected lesions is obtained based on the calculated Dice index value.
在一个具体的示例中，未正确检测的病灶数量按照如下方法得到：In a specific example, the number of incorrectly detected lesions is obtained as follows:
在第二病灶分割模型每一次训练结束时，从数据集的标注图像中逐个抽取病灶，将抽取的病灶分别与预测标注图像中的病灶进行Dice指标值计算，将计算得到的Dice指标值记录到二维的第一列表S中，第一列表S的行表示数据集的标注图像中的病灶，第一列表S的列表示预测标注图像中的病灶；At the end of each training of the second lesion segmentation model, lesions are extracted one by one from the annotated image of the data set, and Dice index values are calculated for the extracted lesions and the lesions in the predicted annotated image, respectively, and the calculated Dice index values are recorded in a two-dimensional first list S, where the rows of the first list S represent the lesions in the annotated image of the data set, and the columns of the first list S represent the lesions in the predicted annotated image;
根据Dice指标值的大小对第一列表S的每一列分别进行排序；Sort each column of the first list S according to the value of the Dice index;
针对第一列表S的每一列确定最大的Dice指标值是否不小于第一数值；Determine for each column of the first list S whether the maximum Dice index value is not less than the first value;
若是，说明数据集中医生的标注图像的病灶和预测标注图像的病灶是正确的匹配，则该Dice指标值对应的病灶被第二病灶分割模型正确检测，未正确检测的病灶数量的值不变；否则，该病灶未被第二病灶分割模型正确检测，未正确检测的病灶数量的值加一；If yes, it means that the lesions in the doctor's annotated image in the data set and the lesions in the predicted annotated image are correctly matched, then the lesions corresponding to the Dice index value are correctly detected by the second lesion segmentation model, and the value of the number of lesions that are not correctly detected remains unchanged; otherwise, the lesions are not correctly detected by the second lesion segmentation model, and the value of the number of lesions that are not correctly detected increases by one;
统计医生的标注图像中每一个病灶是否被正确检测后，得到最终的未正确检测的病灶数量的值numneg。After counting whether each lesion in the doctor's annotated image is correctly detected, the final valuenumneg of the number of lesions that are not correctly detected is obtained.
为了适应不同器官的病灶分割任务，这里给出上述未正确检测的病灶损失计算式中a和b参数的设置方法：In order to adapt to the lesion segmentation tasks of different organs, the setting method of thea andb parameters in the above incorrectly detected lesion loss calculation formula is given here:
设置参数时，固定a=10不变，调节b的大小。当a固定为10，而b分别为0.5或5时，该函数的曲线如图3所示。When setting parameters,a is fixed at 10 andb is adjusted. Whena is fixed at 10 andb is 0.5 or 5, the curve of the function is shown in Figure 3.
其中曲线1参数为a=10和b=5，曲线2参数为a=10和b=0.5。从图3中可以看出，b越小，前期曲线越平缓，后期曲线越陡峭，由未正确检测的病灶引发的loss值就越小，同时未正确检测的病灶数量与loss值的变化也越均匀。b越大，前期曲线越陡峭，后期越平缓，由未正确检测的病灶引发的loss值就越大，同时也难以保证未正确检测的病灶数量与loss值的变化是均匀的。The parameters of curve 1 area = 10 andb = 5, and the parameters of curve 2 area = 10 andb = 0.5. As can be seen from Figure 3, the smallerb is, the flatter the curve in the early stage, the steeper the curve in the later stage, the smaller the loss value caused by incorrectly detected lesions, and the more uniform the changes in the number of incorrectly detected lesions and the loss value. The largerb is, the steeper the curve in the early stage, the flatter the curve in the later stage, the larger the loss value caused by incorrectly detected lesions, and it is difficult to ensure that the changes in the number of incorrectly detected lesions and the loss value are uniform.
数据集中包括多个CT数据，以文件夹的形式存储。每一个CT数据的文件夹中包含一例CT数据的压缩包及对应的标注图像的压缩包。在一些实施方式中，读取和预处理数据集，包括：The data set includes multiple CT data stored in the form of folders. Each CT data folder contains a compressed package of a CT data and a compressed package of a corresponding annotated image. In some embodiments, reading and preprocessing the data set includes:
步骤S11a、打乱数据集中各CT数据的顺序并读取打乱顺序后的各CT数据，以避免数据偏差。Step S11a, disrupting the order of each CT data in the data set and reading each CT data after the disrupted order to avoid data deviation.
步骤S11b、将读取的各CT数据中的CT图像对应标注图像中的病灶标注进行膨胀运算。例如，膨胀运算为核为3的膨胀运算。Step S11b: Perform a dilation operation on the CT image in each of the read CT data corresponding to the lesion annotation in the annotation image. For example, the dilation operation is a dilation operation with a kernel of 3.
步骤S11c、为了利于后续训练，将每一例CT图像及其标注图像的方向转换为RAI（Right Anterior Inferior，右前下）方向。RAI方向是解剖学坐标系中的概念，该坐标系是以人体特征建立起来的。在解剖学坐标系中，S（Superior）与I（Inferior）是一对相反的方向，分别代表上侧（头侧）和下侧（足侧）；L（Left）与R（Right）是一对相反的方向，分别代表左侧和右侧；A（Anterior）与P（Posterior）是一对相反的方向，分别代表前侧与后侧。RAI方向即是右侧、前侧与下侧构成空间坐标系的三个正方向。Step S11c, in order to facilitate subsequent training, the direction of each CT image and its annotated image is converted to the RAI (Right Anterior Inferior) direction. The RAI direction is a concept in the anatomical coordinate system, which is established based on the characteristics of the human body. In the anatomical coordinate system, S (Superior) and I (Inferior) are a pair of opposite directions, representing the upper side (head side) and the lower side (foot side); L (Left) and R (Right) are a pair of opposite directions, representing the left side and the right side respectively; A (Anterior) and P (Posterior) are a pair of opposite directions, representing the front side and the back side respectively. The RAI direction is the three positive directions of the right side, the front side and the bottom side that constitute the spatial coordinate system.
步骤S11d、对每一例CT图像，进行窗宽窗位的调整。如果要分割的病灶为肺结节，则需要调整为肺窗，如果要分割的病灶为肝脏病灶，则需要调整为腹窗。Step S11d: For each CT image, adjust the window width and window position. If the lesion to be segmented is a lung nodule, it needs to be adjusted to a lung window; if the lesion to be segmented is a liver lesion, it needs to be adjusted to an abdominal window.
步骤S11e、对每一例CT图像进行像素归一化处理，将图像所有像素归一化到0~1的范围内。Step S11e: perform pixel normalization processing on each CT image, and normalize all pixels of the image to a range of 0 to 1.
步骤S11f、将每一例CT图像及对应标注图像，根据感兴趣区域进行三维裁剪。如果要分割的病灶为肺结节，则肺部区域为感兴趣区域，如果要分割的病灶为肝脏病灶，则肝脏区域为感兴趣区域。Step S11f, each CT image and the corresponding annotated image is three-dimensionally cropped according to the region of interest. If the lesion to be segmented is a lung nodule, the lung region is the region of interest; if the lesion to be segmented is a liver lesion, the liver region is the region of interest.
步骤S11g、将裁剪后的CT图像及对应标注图像的X轴和Y轴进行尺度缩放，Z轴不变。例如，可将裁剪后的CT图像以及标注图像放大至z*512*512的像素尺寸。Step S11g: scaling the X-axis and Y-axis of the cropped CT image and the corresponding annotated image, and keeping the Z-axis unchanged. For example, the cropped CT image and the annotated image may be enlarged to a pixel size of z*512*512.
步骤S11h、将每一例CT图像及对应标注图像进行切片处理，得到一组切片图像。在一个示例中，将每一例CT图像及对应标注图像切分为8*512*512或16*512*512的切片图像。Step S11h: Slice each CT image and the corresponding annotated image to obtain a set of slice images. In one example, each CT image and the corresponding annotated image is sliced into 8*512*512 or 16*512*512 slice images.
步骤S11i、将各CT数据对应的所有组切片图像打乱顺序，得到读取和预处理的结果。Step S11i: shuffle the order of all the group slice images corresponding to each CT data to obtain the reading and preprocessing results.
在一些实施例中，基于读取和预处理的结果、第一损失函数和第二损失函数，训练第一病灶分割模型和第二病灶分割模型，训练过程如下：In some embodiments, based on the results of reading and preprocessing, the first loss function and the second loss function, the first lesion segmentation model and the second lesion segmentation model are trained, and the training process is as follows:
步骤S12a、读取和预处理的结果中按批次抽取图像数据进行在线数据增强处理。读取和预处理的结果中包含对于一组一组切片图像整合的结果，从中抽取批次数据，输入模型进行训练。将一个批次数据输入模型进行训练时，首先进行在线数据增强处理。 在一些示例中，数据增强处理中的运动模糊的概率为0.04~0.1，图像均值滤波的概率为0.01~0.5，图像水平翻转的概率为0.1~0.3，图像垂直翻转的概率为0.1~0.3，图像的随机平移、尺寸和旋转变换的概率是0.2~0.4，图像产生网格畸变、光学畸变或弹性变换的概率是0.1~0.3，图像产生高斯噪声的概率为0.05~0.1。Step S12a, extract image data in batches from the results of reading and preprocessing for online data enhancement processing. The results of reading and preprocessing include the results of integrating a group of slice images, from which batch data are extracted and input into the model for training. When a batch of data is input into the model for training, online data enhancement processing is first performed. In some examples, the probability of motion blur in data enhancement processing is 0.04~0.1, the probability of image mean filtering is 0.01~0.5, the probability of image horizontal flipping is 0.1~0.3, the probability of image vertical flipping is 0.1~0.3, the probability of random translation, size and rotation transformation of the image is 0.2~0.4, the probability of image grid distortion, optical distortion or elastic transformation is 0.1~0.3, and the probability of image Gaussian noise is 0.05~0.1.
步骤S12b、基于在线数据增强后的图像数据进行第一病灶分割模型或第二病灶分割模型的训练，并计算当前轮中每一次的损失函数值，i表示第i次训练。Step S12b: training the first lesion segmentation model or the second lesion segmentation model based on the image data after online data enhancement, and calculating the loss function value of each time in the current round , i represents the i-th training.
需要说明的是，在训练第一病灶分割模型时，损失函数为第一损失函数。在训练第二病灶分割模型时，前I轮训练的损失函数为第一损失函数，后J轮训练的损失函数为第二损失函数。It should be noted that when training the first lesion segmentation model, the loss function is the first loss function When training the second lesion segmentation model, the loss function of the first round of training is the first loss function , the loss function of the next J rounds of training is the second loss function .
步骤S12c、计算当前平均损失函数值，计算方法：损失函数值之和除以训练次数，计算式如下：Step S12c, calculate the current average loss function value, the calculation method is: the sum of the loss function values divided by the number of training times, the calculation formula is as follows:
其中，是第i次的损失函数值，i为第i次训练，n为训练次数。in, is the i-th loss function value, i is the i-th training, and n is the number of training times.
当训练第一病灶分割模型时，恒为。当训练第二病灶分割模型时，前I轮训练的为，后J轮训练的为。其中I的取值范围是1~20，优选的，I=10；J的取值范围是1~100，优选的，J=50。When training the first lesion segmentation model, Hengwei When training the second lesion segmentation model, the first round of training for , after the J round of training for The value range of I is 1-20, preferably, I=10; the value range of J is 1-100, preferably, J=50.
步骤S12d、计算当前轮中每一次的损失函数值与当前平均损失函数值的大小：若当前轮中每一次的损失函数值≥当前平均损失函数值，则将当前批次图像数据（图像和对应的标注图像）存储到第二列表N中。Step S12d: Calculate the loss function value for each round and the current average loss function value Size: If the loss function value of each time in the current round is ≥ the current average loss function value, the current batch of image data (images and corresponding labeled images) is stored in the second list N.
步骤S12e、当前轮图像数据训练结束后，读取第二列表N中图像数据逐个进行补充训练，若第二列表N中任一批次图像数据对应的损失函数值≥当前平均损失函数值，则将该批次图像数据存储到第三列表M中。Step S12e: After the current round of image data training is completed, read the image data in the second list N one by one for supplementary training. If the loss function value corresponding to any batch of image data in the second list N is ≥ the current average loss function value, then store the batch of image data in the third list M.
步骤S12f、每一轮训练结束后，将第二列表中的内容更新为第三列表中的内容。Step S12f: After each round of training, the content in the second list is updated to the content in the third list.
假设表示第二列表N中的第i个批次的图像和对应的标注图像，代表列表N中第i个批次图像分割结果的损失值。那么当时，将批次的图像和对应的标注图像存储到第三列表M中；当时，不会将当前批次图像和对应的标注图像存储到列表M中。最后将列表N中的内容更新为列表M中的内容。M清空。转至步骤S12a进行下一批次的训练。Assumptions represents the i-th batch of images and the corresponding labeled images in the second list N, represents the loss value of the i-th batch image segmentation result in list N. Then when When The batch of images and the corresponding annotated images are stored in the third list M; , the current batch of images and the corresponding annotated images will not be stored in list M. Finally, the content in list N is updated to the content in list M. M is cleared. Go to step S12a to perform the next batch of training.
步骤S12g、判断当前平均损失函数值是否大于预设阈值：若是，则继续进行下一轮训练；否则结束训练，保存平均损失函数值最小时的第一病灶分割模型或第二病灶分割模型。Step S12g: Determine the current average loss function value Is it greater than a preset threshold: If so, continue with the next round of training; otherwise, end the training and save the first lesion segmentation model or the second lesion segmentation model when the average loss function value is the smallest.
在一些实施例中，读取和预处理数据集之前，还包括：对数据集中的CT数据进行离线数据增强，以对数据集的CT数据扩增。In some embodiments, before reading and preprocessing the data set, the method further includes: performing offline data enhancement on the CT data in the data set to amplify the CT data in the data set.
在肺结节分割场景中，磨玻璃结节与大多实性结节颜色和纹理有所区别，而且数量稀少，不易训练。此情况下，可以优先使用离线数据增强技术，对磨玻璃结节进行数据扩增。具体方法为：根据标注图像，将若干不同CT图像中的磨玻璃肺结节区域裁剪和提取出来，记录在一个列表A之中。再将全部正常无肺结节的CT图像抽取出来，记录在一个列表B之中。随机抽取列表B中的一个CT图像b，并找出该CT图像b的肺部区域。随机可重复抽取列表A中的若干个（可以是1~4个）磨玻璃结节区域图像，对这些磨玻璃结节图像做3D随机旋转变换后，磨玻璃肺结节以0.8~0.9的权重，将磨玻璃结节区域图像融合到CT图像b的肺部区域内的随机位置，得到一个包含磨玻璃肺结节病灶的CT图像c。以此类推，将B列表中全部CT图像合成完毕后，加入到预处理前的数据集中，即可完成离线数据增强。In the pulmonary nodule segmentation scenario, ground glass nodules are different from most solid nodules in color and texture, and their number is rare and difficult to train. In this case, offline data enhancement technology can be used first to amplify the data of ground glass nodules. The specific method is: according to the annotated image, the ground glass lung nodule area in several different CT images is cropped and extracted, and recorded in a list A. Then all normal CT images without lung nodules are extracted and recorded in a list B. A CT image b is randomly extracted from list B, and the lung area of the CT image b is found. Several (can be 1 to 4) ground glass nodule area images in list A are randomly and repeatedly extracted. After performing 3D random rotation transformation on these ground glass nodule images, the ground glass lung nodules are fused to the random position in the lung area of CT image b with a weight of 0.8 to 0.9, and a CT image c containing ground glass lung nodule lesions is obtained. Similarly, after all CT images in list B are synthesized, they are added to the data set before preprocessing to complete offline data enhancement.
获得第一病灶分割模型和第二病灶分割模型，即可用于模型部署，以通过第一病灶分割模型和第二病灶分割模型分割CT图像中的病灶。The first lesion segmentation model and the second lesion segmentation model are obtained, which can be used for model deployment to segment lesions in CT images using the first lesion segmentation model and the second lesion segmentation model.
实施例二Embodiment 2
图4为本公开实施例提供的一种病灶分割方法的流程示意图。如图4所示，一种病灶分割方法，包括：FIG4 is a flow chart of a lesion segmentation method provided by an embodiment of the present disclosure. As shown in FIG4 , a lesion segmentation method includes:
步骤S21、获取CT图像。Step S21, acquiring a CT image.
在实际应用中，获取CT图像，将CT图像进行方向转换，全部转换为RAI方向。对CT图像进行窗宽窗位的调整。如果要分割的病灶为肺结节，则调整为肺窗，如果要分割的是肝脏病灶，则调整为腹窗。对CT图像进行像素归一化处理，将图像所有像素归一化到0~1的范围内。找出CT图像的感兴趣区域进行三维裁剪。如果要分割的是肺结节，则肺部区域为感兴趣区域，如果要分割的是肝脏病灶，则肝脏区域为感兴趣区域。将裁剪后的CT图像，对其X轴和Y轴重新进行尺度缩放，Z轴不变。例如，可将裁剪后的CT图像放大至z*512*512的像素尺寸。In practical applications, a CT image is acquired, and the direction of the CT image is converted to the RAI direction. The window width and window position of the CT image are adjusted. If the lesion to be segmented is a lung nodule, it is adjusted to a lung window. If the lesion to be segmented is a liver lesion, it is adjusted to an abdominal window. The CT image is pixel normalized, and all pixels of the image are normalized to the range of 0 to 1. The region of interest of the CT image is found for three-dimensional cropping. If the lesion to be segmented is a lung nodule, the lung region is the region of interest. If the lesion to be segmented is a liver lesion, the liver region is the region of interest. The cropped CT image is rescaled on the X and Y axes, and the Z axis remains unchanged. For example, the cropped CT image can be enlarged to a pixel size of z*512*512.
步骤S22、将CT图像进行切片，得到一组切片图像。例如将其切分为8*512*512或16*512*512的切片图像，作为CT图像数据读取和预处理的结果。Step S22, slicing the CT image to obtain a set of slice images, for example, slicing the CT image into 8*512*512 or 16*512*512 slice images as the result of CT image data reading and preprocessing.
步骤S23、将该组切片图像依次放入第一病灶分割模型和第二病灶分割模型中进行计算，得到第一病灶分割模型分割结果和第二病灶分割模型分割结果，第一病灶分割模型和第二病灶分割模型是基于实施例一所述的病灶分割模型的训练方法训练得到的。Step S23, sequentially placing the group of slice images into the first lesion segmentation model and the second lesion segmentation model for calculation to obtain the segmentation results of the first lesion segmentation model and the second lesion segmentation model, wherein the first lesion segmentation model and the second lesion segmentation model are trained based on the training method of the lesion segmentation model described in Example 1.
在一些实施例中，病灶分割方法还包括：In some embodiments, the lesion segmentation method further comprises:
步骤S24、对第一病灶分割模型分割结果和第二病灶分割模型分割结果求并集或者交集。Step S24: finding the union or intersection of the segmentation result of the first lesion segmentation model and the segmentation result of the second lesion segmentation model.
在实际应用中，可以对两种模型的分割结果求并集，从而提高病灶检出率，尤其提高小尺寸病灶的检出率。在一些情况下，也可以对两种模型的分割结果求交集，获得置信度相对较高的分割结果，从而减少病灶的误检个数。In practical applications, the segmentation results of the two models can be combined to improve the detection rate of lesions, especially the detection rate of small lesions. In some cases, the segmentation results of the two models can also be intersected to obtain a segmentation result with relatively high confidence, thereby reducing the number of false lesions.
实施例三Embodiment 3
本实施例提供了一种病灶分割模型的训练装置，包括：This embodiment provides a training device for a lesion segmentation model, including:
数据处理模块，用于读取和预处理数据集，数据集包含CT数据，CT数据包括CT图像及对应的标注图像；A data processing module, used for reading and preprocessing a data set, the data set includes CT data, and the CT data includes a CT image and a corresponding annotated image;
模型训练模块，用于基于读取和预处理的结果、第一损失函数和第二损失函数，训练第一病灶分割模型和第二病灶分割模型；其中，第一病灶分割模型和第二病灶分割模型均为神经网络模型，第一病灶分割模型的训练使用第一损失函数，第二病灶分割模型的前预设轮训练使用第一损失函数，第二病灶分割模型的其他轮训练使用第二损失函数，以使训练得到的第二病灶分割模型分割的病灶尺寸小于训练得到的第一病灶分割模型分割的病灶尺寸。A model training module is used to train a first lesion segmentation model and a second lesion segmentation model based on the results of reading and preprocessing, a first loss function and a second loss function; wherein the first lesion segmentation model and the second lesion segmentation model are both neural network models, the first lesion segmentation model is trained using the first loss function, the first preset round of training of the second lesion segmentation model uses the first loss function, and other rounds of training of the second lesion segmentation model use the second loss function, so that the size of the lesion segmented by the trained second lesion segmentation model is smaller than the size of the lesion segmented by the trained first lesion segmentation model.
在一些实施方式中，第二损失函数表达式如下：In some implementations, the second loss function is expressed as follows:
其中，表示第i个病灶的损失权重，的数值基于第i个病灶占有的像素数量确定，第i个病灶占有的像素数量越少，的数值越大；表示第i个病灶的交叉熵损失；表示第i个病灶的Tversky损失；a0和b0分别表示和的系数；表示未正确检测的病灶损失，未正确检测的病灶越多，的数值越大。a0和b0的取值范围是0.0~1.0，考虑到病灶本身尺寸较小，在保证模型稳定训练的同时，也需要保证病灶的检出率，设置a0为0.7，保证模型稳定训练，设置b0为0.5，以达到提高病灶检出率的要求。in, represents the loss weight of thei- th lesion, The value of is determined based on the number of pixels occupied by the i- th lesion. The smaller the number of pixels occupied by the i- th lesion, The larger the value of; represents the cross entropy loss of thei- th lesion; represents the Tversky loss of theith lesion;a 0 andb 0 represent and The coefficient of Indicates the loss of incorrectly detected lesions. The more incorrectly detected lesions are, The larger the value ofa0 andb0 is, the range of a0 and b0 is 0.0~1.0. Considering the small size of the lesion itself, it is necessary to ensure the detection rate of the lesion while ensuring the stable training of the model.a0 is set to 0.7 to ensure the stable training of the model, andb0 is set to 0.5 to achieve the requirement of improving the detection rate of the lesion.
在一些实施例中，通过如下计算式确定第i个病灶的损失权重：In some embodiments, the loss weight of theith lesion is determined by the following calculation formula:
其中，a1和b1均表示可调节的参数，通过调节a1和b1能够控制第i个病灶的损失权重；表示第i个病灶占有的像素数量，表示所有病灶占有的像素数量。Among them,a 1 andb 1 are adjustable parameters. By adjustinga 1 andb 1, the loss weight of thei- th lesion can be controlled; represents the number of pixels occupied by thei -th lesion, Indicates the number of pixels occupied by all lesions.
在一些实施方式中，通过如下计算式确定未正确检测的病灶损失：In some embodiments, incorrectly detected lesion loss is determined by the following calculation:
其中，a和b均表示可调节的参数，通过调节a和b能够控制未正确检测的病灶的惩罚大小；numneg表示未正确检测的病灶数量。通过在第二损失函数中采用第i个病灶的损失权重和未正确检测的病灶损失两种控制手段来调节小尺寸病灶的惩罚程度，以控制小尺寸病灶的检出率。Where,a andb are adjustable parameters. By adjustinga andb , the penalty for incorrectly detected lesions can be controlled.numneg represents the number of incorrectly detected lesions. By using the loss weight of theith lesion in the second loss function and incorrectly detected lesion loss Two control measures are used to adjust the degree of penalty for small lesions to control the detection rate of small lesions.
在一些实施方式中，未正确检测的病灶数量可以按照如下方法得到：在第二病灶分割模型每一次训练结束时，计算数据集的标注图像中每一个病灶和预测标注图像中每一个病灶的Dice指标值，基于计算得到的Dice指标值统计得到未正确检测的病灶数量。In some embodiments, the number of incorrectly detected lesions can be obtained as follows: at the end of each training of the second lesion segmentation model, the Dice index value of each lesion in the annotated image of the data set and each lesion in the predicted annotated image is calculated, and the number of incorrectly detected lesions is obtained based on the calculated Dice index value.
在一个具体的示例中，未正确检测的病灶数量按照如下方法得到：In a specific example, the number of incorrectly detected lesions is obtained as follows:
在第二病灶分割模型每一次训练结束时，从数据集的标注图像中逐个抽取病灶，将抽取的病灶分别与预测标注图像中的病灶进行Dice指标值计算，将计算得到的Dice指标值记录到二维的第一列表S中，第一列表S的行表示数据集的标注图像中的病灶，第一列表S的列表示预测标注图像中的病灶；At the end of each training of the second lesion segmentation model, lesions are extracted one by one from the annotated image of the data set, and Dice index values are calculated for the extracted lesions and the lesions in the predicted annotated image, respectively, and the calculated Dice index values are recorded in a two-dimensional first list S, where the rows of the first list S represent the lesions in the annotated image of the data set, and the columns of the first list S represent the lesions in the predicted annotated image;
根据Dice指标值的大小对第一列表S的每一列分别进行排序；Sort each column of the first list S according to the value of the Dice index;
针对第一列表S的每一列确定最大的Dice指标值是否不小于第一数值；Determine for each column of the first list S whether the maximum Dice index value is not less than the first value;
若是，说明数据集中医生的标注图像的病灶和预测标注图像的病灶是正确的匹配，则该Dice指标值对应的病灶被第二病灶分割模型正确检测，未正确检测的病灶数量的值不变；否则，该病灶未被第二病灶分割模型正确检测，未正确检测的病灶数量的值加一；If yes, it means that the lesions in the doctor's annotated image in the data set and the lesions in the predicted annotated image are correctly matched, then the lesions corresponding to the Dice index value are correctly detected by the second lesion segmentation model, and the value of the number of lesions that are not correctly detected remains unchanged; otherwise, the lesions are not correctly detected by the second lesion segmentation model, and the value of the number of lesions that are not correctly detected increases by one;
统计医生的标注图像中每一个病灶是否被正确检测后，得到最终的未正确检测的病灶数量的值numneg。After counting whether each lesion in the doctor's annotated image is correctly detected, the final valuenumneg of the number of lesions that are not correctly detected is obtained.
数据集中包括多个CT数据，以文件夹的形式存储。每一个CT数据的文件夹中包含一例CT数据的压缩包及对应的标注图像的压缩包。在一些实施方式中，读取和预处理数据集，包括：The data set includes multiple CT data stored in the form of folders. Each CT data folder contains a compressed package of a CT data and a compressed package of a corresponding annotated image. In some embodiments, reading and preprocessing the data set includes:
打乱数据集中各CT数据的顺序并读取打乱顺序后的各CT数据。将读取的各CT数据中的CT图像对应标注图像中的病灶标注进行膨胀运算。将每一例CT图像及其标注图像的方向转换为RAI（Right Anterior Inferior，右前下）方向。对每一例CT图像，进行窗宽窗位的调整。对每一例CT图像进行像素归一化处理，将图像所有像素归一化到0~1的范围内。将每一例CT图像及对应标注图像，根据感兴趣区域进行三维裁剪。将裁剪后的CT图像及对应标注图像的X轴和Y轴进行尺度缩放，Z轴不变。将每一例CT图像及对应标注图像进行切片处理，得到一组切片图像。将各CT数据对应的所有组切片图像打乱顺序，得到读取和预处理的结果。The order of each CT data in the data set is shuffled and each CT data after the shuffle is read. The CT image in each CT data read is dilated according to the lesion annotation in the corresponding annotation image. The direction of each CT image and its annotation image is converted to the RAI (Right Anterior Inferior) direction. For each CT image, the window width and window position are adjusted. Pixel normalization is performed on each CT image, and all pixels of the image are normalized to the range of 0~1. Each CT image and the corresponding annotation image are three-dimensionally cropped according to the region of interest. The X-axis and Y-axis of the cropped CT image and the corresponding annotation image are scaled, and the Z-axis remains unchanged. Each CT image and the corresponding annotation image are sliced to obtain a group of slice images. The order of all groups of slice images corresponding to each CT data is shuffled to obtain the results of reading and preprocessing.
在一些实施例中，基于读取和预处理的结果、第一损失函数和第二损失函数，训练第一病灶分割模型和第二病灶分割模型，训练过程如下：In some embodiments, based on the results of reading and preprocessing, the first loss function and the second loss function, the first lesion segmentation model and the second lesion segmentation model are trained, and the training process is as follows:
读取和预处理的结果中按批次抽取图像数据进行在线数据增强处理。基于在线数据增强后的图像数据进行第一病灶分割模型或第二病灶分割模型的训练，并计算当前轮中每一次的损失函数值。计算当前平均损失函数值。计算当前轮中每一次的损失函数值与当前平均损失函数值的大小：若当前轮中每一次的损失函数值≥当前平均损失函数值，则将当前批次图像数据（图像和对应的标注图像）存储到第二列表N中。当前轮图像数据训练结束后，读取第二列表N中图像数据逐个进行补充训练，若第二列表N中任一批次图像数据对应的损失函数值≥当前平均损失函数值，则将该批次图像数据存储到第三列表M中。每一轮训练结束后，将第二列表中的内容更新为第三列表中的内容。判断当前平均损失函数值是否大于预设阈值：若是，则继续进行下一轮训练；否则结束训练，保存平均损失函数值最小时的第一病灶分割模型或第二病灶分割模型。Extract image data in batches from the results of reading and preprocessing for online data enhancement. Train the first lesion segmentation model or the second lesion segmentation model based on the image data after online data enhancement, and calculate the loss function value for each time in the current round. Calculate the current average loss function value. Calculate the loss function value for each time in the current round and the current average loss function value Size: If the loss function value of each time in the current round is ≥ the current average loss function value, the current batch of image data (images and corresponding annotated images) is stored in the second list N. After the current round of image data training is completed, the image data in the second list N are read one by one for supplementary training. If the loss function value corresponding to any batch of image data in the second list N is ≥ the current average loss function value, the batch of image data is stored in the third list M. After each round of training, the content in the second list is updated to the content in the third list. Determine the current average loss function value Is it greater than a preset threshold: If so, continue with the next round of training; otherwise, end the training and save the first lesion segmentation model or the second lesion segmentation model when the average loss function value is the smallest.
在一些实施例中，读取和预处理数据集之前，还包括：对数据集中的CT数据进行离线数据增强，以对数据集的CT数据扩增。In some embodiments, before reading and preprocessing the data set, the method further includes: performing offline data enhancement on the CT data in the data set to amplify the CT data in the data set.
在肺结节分割场景中，磨玻璃结节与大多实性结节颜色和纹理有所区别，而且数量稀少，不易训练。此情况下，可以优先使用离线数据增强技术，对磨玻璃结节进行数据扩增。具体方法为：根据标注图像，将若干不同CT图像中的磨玻璃肺结节区域裁剪和提取出来，记录在一个列表A之中。再将全部正常无肺结节的CT图像抽取出来，记录在一个列表B之中。随机抽取列表B中的一个CT图像b，并找出该CT图像b的肺部区域。随机可重复抽取列表A中的若干个（可以是1~4个）磨玻璃结节区域图像，对这些磨玻璃结节图像做3D随机旋转变换后，磨玻璃肺结节以0.8~0.9的权重，将磨玻璃结节区域图像融合到CT图像b的肺部区域内的随机位置，得到一个包含磨玻璃肺结节病灶的CT图像c。以此类推，将B列表中全部CT图像合成完毕后，加入到预处理前的数据集中，即可完成离线数据增强。In the pulmonary nodule segmentation scenario, ground glass nodules are different from most solid nodules in color and texture, and their number is rare and difficult to train. In this case, offline data enhancement technology can be used first to amplify the data of ground glass nodules. The specific method is: according to the annotated image, the ground glass lung nodule area in several different CT images is cropped and extracted, and recorded in a list A. Then all normal CT images without lung nodules are extracted and recorded in a list B. A CT image b is randomly extracted from list B, and the lung area of the CT image b is found. Several (can be 1 to 4) ground glass nodule area images in list A are randomly and repeatedly extracted. After performing 3D random rotation transformation on these ground glass nodule images, the ground glass lung nodules are fused to the random position in the lung area of CT image b with a weight of 0.8 to 0.9, and a CT image c containing ground glass lung nodule lesions is obtained. Similarly, after all CT images in list B are synthesized, they are added to the data set before preprocessing to complete offline data enhancement.
获得第一病灶分割模型和第二病灶分割模型，即可用于模型部署，以通过第一病灶分割模型和第二病灶分割模型分割CT图像中的病灶。The first lesion segmentation model and the second lesion segmentation model are obtained, which can be used for model deployment to segment lesions in CT images using the first lesion segmentation model and the second lesion segmentation model.
本实施例至少具备实施例一的全部技术效果，此处不再赘述。This embodiment has at least all the technical effects of the first embodiment, which will not be described in detail here.
实施例四Embodiment 4
本实施例提供了一种病灶分割装置，包括：This embodiment provides a lesion segmentation device, including:
图像获取模块，用于获取CT图像；An image acquisition module, used for acquiring CT images;
图像切片模块，用于将CT图像进行切片，得到一组切片图像；An image slicing module is used to slice the CT image to obtain a set of slice images;
病灶分割模块，用于将该组切片图像依次放入第一病灶分割模型和第二病灶分割模型中进行计算，得到第一病灶分割模型分割结果和第二病灶分割模型分割结果，第一病灶分割模型和第二病灶分割模型是基于实施例三所述病灶分割模型的训练装置训练得到的。The lesion segmentation module is used to sequentially place the group of slice images into the first lesion segmentation model and the second lesion segmentation model for calculation to obtain the segmentation results of the first lesion segmentation model and the second lesion segmentation model. The first lesion segmentation model and the second lesion segmentation model are trained by the training device of the lesion segmentation model described in Example 3.
在一些实施例中，病灶分割模块还对第一病灶分割模型分割结果和第二病灶分割模型分割结果求并集或者交集。In some embodiments, the lesion segmentation module further calculates the union or intersection of the segmentation result of the first lesion segmentation model and the segmentation result of the second lesion segmentation model.
本实施例至少具备实施例二的全部技术效果，此处不再赘述。This embodiment has at least all the technical effects of the second embodiment, which will not be described in detail here.
实施例五Embodiment 5
在上述实施例的基础上，本实施例提供一种计算机设备，包括存储器、处理器及存储在存储器上的计算机程序，处理器执行计算机程序以实现上述实施例所述方法的步骤。On the basis of the above embodiments, this embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory, and the processor executes the computer program to implement the steps of the method described in the above embodiments.
本实施例的一些实施方式中，提供一种计算机可读存储介质，其上存储有计算机程序，该计算机程序被处理器执行时实现上述实施例所述方法的步骤。In some implementations of this embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described in the above embodiment are implemented.
本实施例的一些实施方式中，提供一种计算机程序产品，包括计算机程序/指令，该计算机程序被处理器执行时实现上述实施例所述方法的步骤。In some implementations of this embodiment, a computer program product is provided, including a computer program/instructions, and when the computer program is executed by a processor, the steps of the method described in the above embodiment are implemented.
处理器可以包括但不限于例如一个或者多个处理器或者或微处理器等。每一处理器可以是专用集成电路（Application Specific Integrated Circuit，简称ASIC）、数字信号处理器（Digital Signal Processor，简称DSP）、数字信号处理设备（Digital SignalProcessing Device，简称DSPD）、可编程逻辑器件（Programmable Logic Device，简称PLD）、现场可编程门阵列（Field Programmable Gate Array，简称FPGA）、控制器、微控制器、微处理器或其他电子元件实现，用于执行上述实施例中的方法The processor may include, but is not limited to, one or more processors or microprocessors. Each processor may be an application specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a controller, a microcontroller, a microprocessor or other electronic components to execute the method in the above embodiments.
计算机可读存储介质可以由任何类型的易失性或非易失性存储设备或者它们的组合实现，计算机可读存储介质可以包括但不限于例如，随机存取存储器(RAM)、只读存储器(ROM)、快闪存储器、EPROM存储器、EEPROM存储器、寄存器、计算机存储介质(例如硬碟、软碟、固态硬盘、可移动碟、CD-ROM、 DVD-ROM、蓝光盘等)。Computer-readable storage media can be implemented by any type of volatile or non-volatile storage device or a combination thereof. Computer-readable storage media may include, but are not limited to, for example, random access memory (RAM), read-only memory (ROM), flash memory, EPROM memory, EEPROM memory, registers, computer storage media (such as hard disks, floppy disks, solid-state drives, removable disks, CD-ROMs, DVD-ROMs, Blu-ray disks, etc.).
计算机可读存储介质还可以存储至少一个计算机可执行程序/指令，计算机可执行程序/指令例如是计算机可读指令。计算机可读存储介质包括但不限于例如易失性存储器和/或非易失性存储器。易失性存储器例如可以包括随机存取存储器(RAM)和/或高速缓冲存储器(cache)等。计算机可读存储介质例如可以包括只读存储器(ROM)、硬盘、闪存等。例如，非暂时性计算机可读存储介质可以连接于诸如计算机等的计算设备，接着，在计算设备运行计算机可读存储介质上存储的计算机可读指令的情况下，可以进行如上描述的各个方法。The computer-readable storage medium may also store at least one computer executable program/instruction, which may be, for example, a computer-readable instruction. The computer-readable storage medium includes, but is not limited to, for example, a volatile memory and/or a non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and/or a cache memory (cache), etc. The computer-readable storage medium may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc. For example, a non-transitory computer-readable storage medium may be connected to a computing device such as a computer, and then, when the computing device runs the computer-readable instructions stored on the computer-readable storage medium, the various methods described above may be performed.
除此之外，该计算机设备还可以包括(但不限于)数据总线、输入/输出(I/O)总线，显示器以及输入/输出设备 (例如，键盘、鼠标、扬声器等)等。In addition, the computer device may also include (but not limited to) a data bus, an input/output (I/O) bus, a display, and input/output devices (e.g., keyboard, mouse, speakers, etc.), etc.
处理器可以通过I/O总线经由有线或无线网络与外部设备通信。The processor may communicate with external devices via an I/O bus via a wired or wireless network.
在一个实施方式中，该至少一个计算机可执行指令也可以被编译为或组成一种软件产品/计算机程序产品，其中一个或多个计算机可执行指令被处理器运行时执行本技术所描述的实施例中的各个功能和/或方法的步骤。In one embodiment, the at least one computer executable instruction may also be compiled into or constitute a software product/computer program product, wherein one or more computer executable instructions are executed by a processor to perform the various functions and/or method steps in the embodiments described in the present technology.
在本公开所提供的实施例中，应该理解到，所揭露的装置和方法，也可以通过其它的方式实现。以上所描述的装置实施例仅仅是示意性的，例如，附图中的流程图和框图显示了根据本公开的多个实施例的装置、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段或代码的一部分，上述模块、程序段或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现方式中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个连续的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或动作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。In the embodiments provided in the present disclosure, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the above-mentioned module, program segment or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and/or flowchart, and the combination of boxes in the block diagram and/or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.
需要说明的是，在本公开中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限的要素，并不排除在包括要素的过程、方法、物品或者设备中还存在另外的相同要素。It should be noted that in the present disclosure, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element limited by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.
虽然本公开所揭露的实施方式如上，但上述的内容只是为了便于理解本公开而采用的实施方式，并非用以限定本公开。任何本公开所属技术领域内的技术人员，在不脱离本公开所揭露的精神和范围的前提下，可以在实施的形式上及细节上作任何的修改与变化，但本公开的专利保护范围，仍须以所附的权利要求书所界定的范围为准。Although the embodiments disclosed in the present disclosure are as above, the above contents are only embodiments adopted for facilitating the understanding of the present disclosure and are not intended to limit the present disclosure. Any technician in the technical field to which the present disclosure belongs can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in the present disclosure, but the scope of patent protection of the present disclosure shall still be subject to the scope defined in the attached claims.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411112070.6ACN118628871B (en) | 2024-08-14 | 2024-08-14 | Training method, device, equipment and storage medium of focus segmentation model |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411112070.6ACN118628871B (en) | 2024-08-14 | 2024-08-14 | Training method, device, equipment and storage medium of focus segmentation model |
| Publication Number | Publication Date |
|---|---|
| CN118628871A CN118628871A (en) | 2024-09-10 |
| CN118628871Btrue CN118628871B (en) | 2024-11-01 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411112070.6AActiveCN118628871B (en) | 2024-08-14 | 2024-08-14 | Training method, device, equipment and storage medium of focus segmentation model |
| Country | Link |
|---|---|
| CN (1) | CN118628871B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116503607A (en)* | 2023-06-28 | 2023-07-28 | 天津市中西医结合医院（天津市南开医院） | CT image segmentation method and system based on deep learning |
| CN117611596A (en)* | 2023-12-05 | 2024-02-27 | 桂林电子科技大学 | Semi-supervised lung cancer medical image segmentation method and device based on LesionMix and entropy minimization |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109242860B (en)* | 2018-08-21 | 2021-10-08 | 电子科技大学 | Brain tumor image segmentation method based on deep learning and weight space integration |
| CN109829894B (en)* | 2019-01-09 | 2022-04-26 | 平安科技（深圳）有限公司 | Segmentation model training method, OCT image segmentation method, device, equipment and medium |
| US20220076133A1 (en)* | 2020-09-04 | 2022-03-10 | Nvidia Corporation | Global federated training for neural networks |
| KR102854968B1 (en)* | 2022-03-25 | 2025-09-04 | 재단법인 아산사회복지재단 | Device and method for classifying and segmenting ct image based on machine learning model |
| CN115346076B (en)* | 2022-10-18 | 2023-01-17 | 安翰科技(武汉)股份有限公司 | Pathological image recognition method, model training method and system thereof, and storage medium |
| KR102680328B1 (en)* | 2023-01-25 | 2024-07-01 | 중앙대학교 산학협력단 | Deep learning model optimization method and apparatus for medical image segmentation |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116503607A (en)* | 2023-06-28 | 2023-07-28 | 天津市中西医结合医院（天津市南开医院） | CT image segmentation method and system based on deep learning |
| CN117611596A (en)* | 2023-12-05 | 2024-02-27 | 桂林电子科技大学 | Semi-supervised lung cancer medical image segmentation method and device based on LesionMix and entropy minimization |
| Publication number | Publication date |
|---|---|
| CN118628871A (en) | 2024-09-10 |
| Publication | Publication Date | Title |
|---|---|---|
| TWI754195B (en) | Image processing method and device, electronic device and computer-readable storage medium | |
| Prajapati et al. | Classification of dental diseases using CNN and transfer learning | |
| CN108921854B (en) | A method and system for labeling irregular lesions in gastrointestinal endoscopy images | |
| CN109754396A (en) | Image registration method and device, computer equipment and storage medium | |
| CN110930318A (en) | A low-dose CT image restoration and denoising method | |
| CN111260701B (en) | Multi-mode retina fundus image registration method and device | |
| CN111932467A (en) | Image processing method and device | |
| CN112150472A (en) | Three-dimensional jaw bone image segmentation method and device based on CBCT (cone beam computed tomography) and terminal equipment | |
| KR102408452B1 (en) | Method for diagnosing joint diseases based on deep learning | |
| Chen et al. | Detection of various dental conditions on dental panoramic radiography using Faster R-CNN | |
| CN118115838B (en) | Medical image segmentation model training method, segmentation method, device and program product | |
| CN111553882A (en) | Method and device for identifying femoral head ischemic necrosis through MR (magnetic resonance) image | |
| CN110189324B (en) | Medical image processing method and processing device | |
| CN110047145B (en) | Tissue deformation simulation system and method based on deep learning and finite element modeling | |
| CN117011318A (en) | Tooth CT image three-dimensional segmentation method, system, equipment and medium | |
| CN114219817A (en) | New Coronary Pneumonia CT Image Segmentation Method and Terminal Equipment | |
| CN111080676B (en) | A Method for Tracking Feature Points of Endoscopic Image Sequences by Online Classification | |
| CN111402221B (en) | Image processing method and device and electronic equipment | |
| CN112966390A (en) | Method and apparatus for processing clothes based on dual three-dimensional distance fields | |
| CN118628871B (en) | Training method, device, equipment and storage medium of focus segmentation model | |
| Liao et al. | Classification of dental teeth X-ray images using a deep learning CNN model | |
| CN113822904A (en) | Image labeling device and method and readable storage medium | |
| CN117315378B (en) | Grading judgment method for pneumoconiosis and related equipment | |
| CN110430416B (en) | Free viewpoint image generation method and device | |
| CN111466933A (en) | A method and system for measuring the range of motion of the spine |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |