{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/cell-detection-in-microscopy-images-with-deep","title":"Cell Detection in Microscopy Images with Deep Convolutional Neural Network and Compressed Sensing","arxiv_id":"1708.03307","date":"2017-08-10","proceeding":null,"authors":["Yao Xue","Nilanjan Ray"],"abstract":"The ability to automatically detect certain types of cells or cellular\nsubunits in microscopy images is of significant interest to a wide range of\nbiomedical research and clinical practices. Cell detection methods have evolved\nfrom employing hand-crafted features to deep learning-based techniques. The\nessential idea of these methods is that their cell classifiers or detectors are\ntrained in the pixel space, where the locations of target cells are labeled. In\nthis paper, we seek a different route and propose a convolutional neural\nnetwork (CNN)-based cell detection method that uses encoding of the output\npixel space. For the cell detection problem, the output space is the sparsely\nlabeled pixel locations indicating cell centers. We employ random projections\nto encode the output space to a compressed vector of fixed dimension. Then, CNN\nregresses this compressed vector from the input pixels. Furthermore, it is\npossible to stably recover sparse cell locations on the output pixel space from\nthe predicted compressed vector using $L_1$-norm optimization. In the past,\noutput space encoding using compressed sensing (CS) has been used in\nconjunction with linear and non-linear predictors. To the best of our\nknowledge, this is the first successful use of CNN with CS-based output space\nencoding. We made substantial experiments on several benchmark datasets, where\nthe proposed CNN + CS framework (referred to as CNNCS) achieved the highest or\nat least top-3 performance in terms of F1-score, compared with other\nstate-of-the-art methods.","url_abs":"http://arxiv.org/abs/1708.03307v3","url_pdf":"http://arxiv.org/pdf/1708.03307v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"cell-detection-in-microscopy-images-with-deep","repo_url":"https://github.com/yaoxuexa/CNNCS","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"mxnet","reach":null},{"paper_slug":"cell-detection-in-microscopy-images-with-deep","repo_url":"https://github.com/isgilman/CrossSection_DeepLearning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"cell-detection-in-microscopy-images-with-deep","repo_url":"https://github.com/kevinkayaks/deep-compression","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"cell-detection","task_name":"Cell Detection"},{"task_slug":"compressed-sensing","task_name":"compressed sensing"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}