{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2023,11,25]],"date-time":"2023-11-25T05:22:42Z","timestamp":1700889762373},"reference-count":35,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2023,11,25]],"date-time":"2023-11-25T00:00:00Z","timestamp":1700870400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,11,25]],"date-time":"2023-11-25T00:00:00Z","timestamp":1700870400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BioData Mining"],"abstract":"<jats:title>Abstract<\/jats:title><jats:sec>\n                <jats:title>Background<\/jats:title>\n                <jats:p>Discrimination between patients affected by inflammatory bowel diseases and healthy controls on the basis of endoscopic imaging is an challenging problem for machine learning models. Such task is used here as the testbed for a novel deep learning classification pipeline, powered by a set of solutions enhancing characterising elements such as reproducibility, interpretability, reduced computational workload, bias-free modeling and careful image preprocessing.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Results<\/jats:title>\n                <jats:p>First, an automatic preprocessing procedure is devised, aimed to remove artifacts from clinical data, feeding then the resulting images to an aggregated per-patient model to mimic the clinicians decision process. The predictions are based on multiple snapshots obtained through resampling, reducing the risk of misleading outcomes by removing the low confidence predictions. Each patient\u2019s outcome is explained by returning the images the prediction is based upon, supporting clinicians in verifying diagnoses without the need for evaluating the full set of endoscopic images. As a major theoretical contribution, quantization is employed to reduce the complexity and the computational cost of the model, allowing its deployment on small power devices with an almost negligible 3% performance degradation. Such quantization procedure holds relevance not only in the context of per-patient models but also for assessing its feasibility in providing real-time support to clinicians even in low-resources environments. The pipeline is demonstrated on a private dataset of endoscopic images of 758 IBD patients and 601 healthy controls, achieving Matthews Correlation Coefficient 0.9 as top performance on test set.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Conclusion<\/jats:title>\n                <jats:p>We highlighted how a comprehensive pre-processing pipeline plays a crucial role in identifying and removing artifacts from data, solving one of the principal challenges encountered when working with clinical data. Furthermore, we constructively showed how it is possible to emulate clinicians decision process and how it offers significant advantages, particularly in terms of explainability and trust within the healthcare context. Last but not least, we proved that quantization can be a useful tool to reduce the time and resources consumption with an acceptable degradation of the model performs. The quantization study proposed in this work points up the potential development of real-time quantized algorithms as valuable tools to support clinicians during endoscopy procedures.<\/jats:p>\n              <\/jats:sec>","DOI":"10.1186\/s13040-023-00350-0","type":"journal-article","created":{"date-parts":[[2023,11,25]],"date-time":"2023-11-25T03:01:40Z","timestamp":1700881300000},"update-policy":"http:\/\/dx.doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Endoscopy-based IBD identification by a quantized deep learning pipeline"],"prefix":"10.1186","volume":"16","author":[{"given":"Massimiliano","family":"Datres","sequence":"first","affiliation":[]},{"given":"Elisa","family":"Paolazzi","sequence":"additional","affiliation":[]},{"given":"Marco","family":"Chierici","sequence":"additional","affiliation":[]},{"given":"Matteo","family":"Pozzi","sequence":"additional","affiliation":[]},{"given":"Antonio","family":"Colangelo","sequence":"additional","affiliation":[]},{"given":"Marcello","family":"Dorian\u00a0Donzella","sequence":"additional","affiliation":[]},{"given":"Giuseppe","family":"Jurman","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2023,11,25]]},"reference":[{"key":"350_CR1","doi-asserted-by":"publisher","unstructured":"Banner R, Nahshan Y, Soudry D. Post training 4-bit quantization of convolutional networks for rapid-deployment. In: Proceedings of Advances in Neural Information Processing Systems 2019. vol. 32(714). 2019. p. 7950\u20137958. https:\/\/doi.org\/10.48550\/arXiv.1810.05723.","DOI":"10.48550\/arXiv.1810.05723"},{"key":"350_CR2","doi-asserted-by":"publisher","unstructured":"Bansal R, Raj G, Choudhury T. Blur image detection using Laplacian operator and Open-CV. In: 2016 International Conference System Modeling & Advancement in Research Trends (SMART). IEEE; 2016. p. 63\u201367. https:\/\/doi.org\/10.1109\/SYSMART.2016.7894491.","DOI":"10.1109\/SYSMART.2016.7894491"},{"issue":"10","key":"350_CR3","doi-asserted-by":"publisher","first-page":"1761","DOI":"10.1097\/MIB.0000000000000103","volume":"20","author":"EI Benchimol","year":"2014","unstructured":"Benchimol EI, Manuel DG, Guttmann A, Nguyen GC, Mojaverian N, Quach P, et al. Changing age demographics of inflammatory bowel disease in Ontario, Canada: a population-based cohort study of epidemiology trends. Inflamm Bowel Dis. 2014;20(10):1761\u20139. https:\/\/doi.org\/10.1097\/MIB.0000000000000103.","journal-title":"Inflamm Bowel Dis."},{"key":"350_CR4","unstructured":"Bengio Y, L\u00e9onard N, Courville A. Estimating or propagating gradients through stochastic neurons for conditional computation. 2013. ArXiv\u00a0preprint\u00a0arXiv:1308.3432.\u00a0https:\/\/arxiv.org\/abs\/1308.3432."},{"key":"350_CR5","first-page":"122","volume":"120","author":"Bradski G The","year":"2000","unstructured":"The Bradski G, Library OpenCV. Dr Dobb\u2019s J Softw Tools. 2000;120:122\u20135.","journal-title":"Dr Dobb\u2019s J Softw Tools."},{"key":"350_CR6","doi-asserted-by":"publisher","unstructured":"Bussola N, Marcolini A, Maggio V, Jurman G, Furlanello C. AI Slipping on Tiles: Data Leakage in Digital Pathology. In: Pattern Recognition. ICPR International Workshops and Challenges. ICPR 2021. vol. 12661 of Lecture Notes in Computer Science. Springer; 2021. p. 167\u2013182. https:\/\/doi.org\/10.1007\/978-3-030-68763-2_13.","DOI":"10.1007\/978-3-030-68763-2_13"},{"key":"350_CR7","doi-asserted-by":"publisher","first-page":"635764","DOI":"10.3389\/fbioe.2021.635764","volume":"9","author":"G Chen","year":"2021","unstructured":"Chen G, Shen J. Artificial intelligence enhances studies on inflammatory bowel disease. Front Bioeng Biotechnol. 2021;9:635764. https:\/\/doi.org\/10.3389\/fbioe.2021.635764.","journal-title":"Front Bioeng Biotechnol."},{"key":"350_CR8","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s12864-019-6413-7","volume":"21","author":"D Chicco","year":"2020","unstructured":"Chicco D, Jurman G. The advantages of the Matthews Correlation Coefficient (MCC) over F1 score and accuracy in binary classification evaluation. BMC Genomics. 2020;21:1\u201313. https:\/\/doi.org\/10.1186\/s12864-019-6413-7.","journal-title":"BMC Genomics."},{"issue":"6","key":"350_CR9","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s12911-022-02043-w","volume":"22","author":"M Chierici","year":"2022","unstructured":"Chierici M, Puica N, Pozzi M, Capistrano A, Donzella MD, Colangelo A, et al. Automatically detecting Crohn\u2019s disease and Ulcerative Colitis from endoscopic imaging. BMC Med Inform Dec Making. 2022;22(6):1\u201311. https:\/\/doi.org\/10.1186\/s12911-022-02043-w.","journal-title":"BMC Med Inform Dec Making."},{"key":"350_CR10","unstructured":"Choi J, Wang Z, Venkataramani S, Chuang PIJ, Srinivasan V, Gopalakrishnan K. Pact: Parameterized clipping activation for quantized neural networks. 2018. ArXiv\u00a0preprint\u00a0arXiv:1805.06085.\u00a0https:\/\/arxiv.org\/abs\/1805.06085."},{"key":"350_CR11","doi-asserted-by":"publisher","unstructured":"Choukroun Y, Kravchik E, Yang F, Kisilev P. Low-bit quantization of neural networks for efficient inference. In: 2019 IEEE\/CVF International Conference on Computer Vision Workshop (ICCVW). IEEE; 2019. p. 3009\u20133018. https:\/\/doi.org\/10.1109\/ICCVW.2019.00363.","DOI":"10.1109\/ICCVW.2019.00363"},{"key":"350_CR12","doi-asserted-by":"publisher","unstructured":"Fiesler E, Choudry A, Caulfield HJ. Weight discretization paradigm for optical neural networks. In: Proceedings of The International Congress on Optical Science and Engineering, 1990, The Hague, Netherlands. vol. 1281. SPIE; 1990. p. 164\u2013173. https:\/\/doi.org\/10.1117\/12.20700.","DOI":"10.1117\/12.20700"},{"issue":"1","key":"350_CR13","doi-asserted-by":"publisher","first-page":"23","DOI":"10.1038\/nrgastro.2011.214","volume":"9","author":"G Fiorino","year":"2012","unstructured":"Fiorino G, Bonifacio C, Malesci A, Balzarini L, Danese S. MRI in Crohn\u2019s disease\u2013current and future clinical applications. Nat Rev Gastroenterol Hepatol. 2012;9(1):23\u201331. https:\/\/doi.org\/10.1038\/nrgastro.2011.214.","journal-title":"Nat Rev Gastroenterol Hepatol."},{"key":"350_CR14","unstructured":"Goodfellow IJ, Shlens J, Szegedy C. Explaining and harnessing adversarial examples. 2014. ArXiv\u00a0preprint\u00a0arXiv:1412.6572."},{"key":"350_CR15","unstructured":"Guilford JP. Psychometric methods.\u00a0New York: McGraw-Hill; 1954."},{"key":"350_CR16","doi-asserted-by":"publisher","first-page":"519","DOI":"10.1007\/s10620-012-2371-5","volume":"58","author":"MD Kappelman","year":"2013","unstructured":"Kappelman MD, Moore KR, Allen JK, Cook SF. Recent trends in the prevalence of Crohn\u2019s disease and ulcerative colitis in a commercially insured US population. Dig Dis Sci. 2013;58:519\u201325. https:\/\/doi.org\/10.1007\/s10620-012-2371-5.","journal-title":"Dig Dis Sci."},{"key":"350_CR17","unstructured":"Kingma DP, Ba J. Adam: A method for stochastic optimization. 2014. ArXiv\u00a0preprint\u00a0arXiv:1412.6980.\u00a0https:\/\/arxiv.org\/abs\/1412.6980."},{"key":"350_CR18","unstructured":"Kokhlikyan N, Miglani V, Alsallakh B, Martin M, Reblitz-Richardson O. Investigating sanity checks for saliency maps with image and text classification. 2021. ArXiv\u00a0preprint\u00a0arXiv:2106.07475.\u00a0https:\/\/arxiv.org\/abs\/2106.07475."},{"key":"350_CR19","unstructured":"Kokhlikyan N, Miglani V, Martin M, Wang E, Alsallakh B, Reynolds J, et\u00a0al. Captum: A unified and generic model interpretability library for Pytorch. 2020. ArXiv\u00a0preprint\u00a0arXiv:2009.07896. https:\/\/arxiv.org\/abs\/2009.07896."},{"key":"350_CR20","unstructured":"Krishnamoorthi R. Quantizing deep convolutional networks for efficient inference: A whitepaper. 2018. ArXiv\u00a0preprint\u00a0arXiv:1806.08342.\u00a0https:\/\/arxiv.org\/abs\/1806.08342."},{"issue":"1","key":"350_CR21","doi-asserted-by":"publisher","first-page":"53","DOI":"10.1109\/72.182695","volume":"4","author":"M Marchesi","year":"1993","unstructured":"Marchesi M, Orlandi G, Piazza F, Uncini A. Fast neural networks without multipliers. IEEE Trans Neural Netw. 1993;4(1):53\u201362. https:\/\/doi.org\/10.1109\/72.182695.","journal-title":"IEEE Trans Neural Netw."},{"key":"350_CR22","doi-asserted-by":"publisher","unstructured":"Mostafa S, Mondal D, Beck MA, Bidinosti CP, Henry CJ, Stavness I. Leveraging Guided Backpropagation to Select Convolutional Neural Networks for Plant Classification. Front Artif Intell. 2022;5:Online. https:\/\/doi.org\/10.3389\/frai.2022.871162.","DOI":"10.3389\/frai.2022.871162"},{"key":"350_CR23","doi-asserted-by":"publisher","unstructured":"Moura V, Almeida V, Santos DBS, Costa N, Sousa LL, Pimentel PC. Mobile Device ECG Classification using quantized Neural Networks. 2020. Research Square. https:\/\/doi.org\/10.21203\/rs.3.rs-41077\/v1.","DOI":"10.21203\/rs.3.rs-41077\/v1"},{"issue":"5","key":"350_CR24","doi-asserted-by":"publisher","first-page":"1670","DOI":"10.1053\/j.gastro.2007.09.001","volume":"133","author":"S Nikolaus","year":"2007","unstructured":"Nikolaus S, Schreiber S. Diagnostics of inflammatory bowel disease. Gastroenterology. 2007;133(5):1670\u201389. https:\/\/doi.org\/10.1053\/j.gastro.2007.09.001.","journal-title":"Gastroenterology."},{"issue":"9","key":"350_CR25","doi-asserted-by":"publisher","first-page":"337","DOI":"10.3345\/cep.2019.00500","volume":"63","author":"S Park","year":"2020","unstructured":"Park S, Kang Y, Koh H, Kim S. Increasing incidence of inflammatory bowel disease in children and adolescents: significance of environmental factors. Clin Exp Pediatr. 2020;63(9):337. https:\/\/doi.org\/10.3345\/cep.2019.00500.","journal-title":"Clin Exp Pediatr."},{"key":"350_CR26","unstructured":"Simonyan K, Vedaldi A, Zisserman A. Deep inside convolutional networks: Visualising image classification models and saliency maps. 2013. ArXiv\u00a0preprint\u00a0arXiv:1312.6034.\u00a0https:\/\/arxiv.org\/abs\/1312.6034."},{"key":"350_CR27","doi-asserted-by":"publisher","unstructured":"Sousa JVM, de\u00a0Almeida VR, Saraiva AA, Santos DBS, Pimentel PMC, de\u00a0Sousa LL. Classification of Pneumonia images on mobile devices with Quantized Neural Network. Res Soc Dev. 2020;9(10):e889108382\u2013e889108382. https:\/\/doi.org\/10.33448\/rsd-v9i10.8382.","DOI":"10.33448\/rsd-v9i10.8382"},{"key":"350_CR28","doi-asserted-by":"crossref","unstructured":"Suvorov R, Logacheva E, Mashikhin A, Remizova A, Ashukha A, Silvestrov A, et\u00a0al. Resolution-robust Large Mask Inpainting with Fourier Convolutions. 2021. ArXiv\u00a0preprint\u00a0arXiv:2109.07161.","DOI":"10.1109\/WACV51458.2022.00323"},{"issue":"8","key":"350_CR29","doi-asserted-by":"publisher","first-page":"2150","DOI":"10.1053\/j.gastro.2020.02.012","volume":"158","author":"K Takenaka","year":"2020","unstructured":"Takenaka K, Ohtsuka K, Fujii T, Negi M, Suzuki K, Shimizu H, et al. Development and validation of a deep neural network for accurate evaluation of endoscopic images from patients with ulcerative colitis. Gastroenterology. 2020;158(8):2150\u20137. https:\/\/doi.org\/10.1053\/j.gastro.2020.02.012.","journal-title":"Gastroenterology."},{"key":"350_CR30","doi-asserted-by":"publisher","unstructured":"Tontini GE, Rimondi A, Vernero M, Neumann H, Vecchi M, Bezzio C, et al. Artificial intelligence in gastrointestinal endoscopy for inflammatory bowel disease: a systematic review and new horizons. Ther Adv Gastroenterol. 2021;14:17562848211017730. https:\/\/doi.org\/10.1177\/17562848211017730.","DOI":"10.1177\/17562848211017730"},{"issue":"1","key":"350_CR31","doi-asserted-by":"publisher","first-page":"21","DOI":"10.3748\/wjg.v21.i1.21","volume":"21","author":"GE Tontini","year":"2015","unstructured":"Tontini GE, Vecchi M, Pastorelli L, Neurath MF, Neumann H. Differential diagnosis in inflammatory bowel disease colitis: state of the art and future perspectives. World J Gastroenterol. 2015;21(1):21. https:\/\/doi.org\/10.3748\/wjg.v21.i1.21.","journal-title":"World J Gastroenterol."},{"issue":"3","key":"350_CR32","doi-asserted-by":"publisher","first-page":"e065186","DOI":"10.1136\/bmjopen-2022-065186","volume":"13","author":"R Wang","year":"2023","unstructured":"Wang R, Li Z, Liu S, Zhang D. Global, regional and national burden of inflammatory bowel disease in 204 countries and territories from 1990 to 2019: a systematic analysis based on the Global Burden of Disease Study 2019. BMJ Open. 2023;13(3):e065186. https:\/\/doi.org\/10.1136\/bmjopen-2022-065186.","journal-title":"BMJ Open."},{"key":"350_CR33","unstructured":"Wu H, Judd P, Zhang X, Isaev M, Micikevicius P. Integer quantization for deep learning inference: Principles and empirical evaluation. 2020. ArXiv\u00a0preprint\u00a0arXiv:2004.09602.\u00a0https:\/\/arxiv.org\/abs\/2004.09602."},{"key":"350_CR34","unstructured":"Wu Y, Kirillov A, Massa F, Lo WY, Girshick R. Detectron2. 2019. https:\/\/github.com\/facebookresearch\/detectron2.\u00a0Accessed 23 Nov."},{"key":"350_CR35","unstructured":"Yin P, Lyu J, Zhang S, Osher S, Qi Y, Xin J. Understanding straight-through estimator in training activation quantized neural nets. 2019. ArXiv\u00a0preprint\u00a0arXiv:1903.05662.\u00a0https:\/\/arxiv.org\/abs\/1903.05662."}],"container-title":["BioData Mining"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13040-023-00350-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s13040-023-00350-0\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13040-023-00350-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,25]],"date-time":"2023-11-25T03:01:47Z","timestamp":1700881307000},"score":1,"resource":{"primary":{"URL":"https:\/\/biodatamining.biomedcentral.com\/articles\/10.1186\/s13040-023-00350-0"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,11,25]]},"references-count":35,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2023,12]]}},"alternative-id":["350"],"URL":"https:\/\/doi.org\/10.1186\/s13040-023-00350-0","relation":{},"ISSN":["1756-0381"],"issn-type":[{"value":"1756-0381","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,11,25]]},"assertion":[{"value":"29 October 2023","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"18 November 2023","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"25 November 2023","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The study that produced the data used for these analyses was conducted in accordance with the Declaration of Helsinki and approved by a local ethics committee in Italy, namely the Istituto Tumori Bari \u201cGiovanni Paolo II\u201d - IRCCS, with the protocol number n. 333 \/ C.E. of 31\/07\/2019.Informed consent to participate was obtained from the study participants.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"The team acquired the permission to access and analyze the clinical patient data by GPI S.p.A. through the Contract 210202009864 between GPI S.p.A. and FBK.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare no competing interests.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"33"}}