“…Another limitation of our systematic review is that it only includes open-source software. Open-source [ 88 ] research software is typically developed by small research groups without financial incentives, who generally collaborate for shorter periods than industry software development teams [ 89 , 90 ]. This may lead to the discontinuation of projects sometime after publication [ 89 ].…”

“…Future pipelines should follow best practices for software engineering [ 92 ] to maximize reproducibility, performance, quality, and user-friendliness. These include the use of workflow managers [ 26 , 93 , 94 ], containerization software [ 25 , 95 ], testing [ 96 ], continuous integration/continuous delivery (CI/CD) [ 95 ], comprehensive documentation, and licensing as free and open-source software (FOSS) [ 88 ]. Among the pipelines reviewed, only two were developed using a workflow manager, and this development occurred as recently as 2022 and 2024.…”

“… 40 , 41 In contrast, open-source software, including CellProfiler, allows image analysis methods and codes not only to be accessible to investigators but also to be interrogated, extended, enhanced, and shared by a wider community, which can solve the majority of the challenges around data management, visualization, and interactivity while permitting others to add new algorithms and functionality for specific tasks. 39 , 42 , 43 CellProfiler is an image analysis software released in 2005 and has since undergone a series of improvements to become one of the most widely used forms of software for pathomics analyses due to its versatility, continuous improvements, and ease of use. 24 , 43 Although some useful features might be ignored by using the inherent software extraction method of CellProfiler, there is no perfect software or method providing features that support all studies; thus, a more suitable software or method can be selected for individual research needs.…”

“…The type of license under which free software is developed differs from that of proprietary software, since companies producing the latter, or patented software, typically withhold all the production, modification, and distribution rights of the source code, and only allow the user or purchaser to use to program in question. We then see that the substantial difference between open-source software and proprietary software lies in the use conditions imposed upon by the type of license, more so than in a functional distinction, or related to a programming code ( de Souza, 2023;Eghbal, 2020;Fortunato & Galassi, 2021). This way, handling of free software implies the use of digital tools which motivate the user to self-understand the use thereof, to conduct well-thought-out actions, and to find alternative solutions to current technological requirements, without the typical restrictions on the use of proprietary tools(Nicoll & Keogh, 2019).…”