{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,13]],"date-time":"2026-04-13T17:10:16Z","timestamp":1776100216988,"version":"3.50.1"},"reference-count":46,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,7,15]],"date-time":"2025-07-15T00:00:00Z","timestamp":1752537600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p>Creating training and testing datasets for machine learning algorithms to measure linear dimensions of organs is a tedious task. There are no universally accepted methods for evaluating outliers or anomalies in such datasets. This can cause errors in machine learning and compromise the quality of end products. The goal of this study is to identify optimal methods for detecting organ anomalies and outliers in medical datasets designed to train and test neural networks in morphometrics.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>A dataset was created containing linear measurements of the spleen obtained from CT scans. Labelling was performed by three radiologists. The total number of studies included in the sample was <jats:italic>N<\/jats:italic>\u202f=\u202f197 patients. Using visual methods (1.5 interquartile range; heat map; boxplot; histogram; scatter plot), machine learning algorithms (Isolation forest; Density-Based Spatial Clustering of Applications with Noise; K-nearest neighbors algorithm; Local outlier factor; One-class support vector machines; EllipticEnvelope; Autoencoders), and mathematical statistics (z-score, Grubb\u2019s test; Rosner\u2019s test).<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>We identified measurement errors, input errors, abnormal size values and non-standard shapes of the organ (sickle-shaped, round, triangular, additional lobules). The most effective methods included visual techniques (including boxplots and histograms) and machine learning algorithms such is OSVM, KNN and autoencoders. A total of 32 outlier anomalies were found.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>Curation of complex morphometric datasets must involve thorough mathematical and clinical analyses. Relying solely on mathematical statistics or machine learning methods appears inadequate.<\/jats:p><\/jats:sec>","DOI":"10.3389\/frai.2025.1607348","type":"journal-article","created":{"date-parts":[[2025,7,15]],"date-time":"2025-07-15T05:25:55Z","timestamp":1752557155000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Outliers and anomalies in training and testing datasets for AI-powered morphometry\u2014evidence from CT scans of the spleen"],"prefix":"10.3389","volume":"8","author":[{"given":"Yuriy","family":"Vasilev","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Anastasia","family":"Pamova","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tatiana","family":"Bobrovskaya","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Anton","family":"Vladzimirskyy","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Olga","family":"Omelyanskaya","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Elena","family":"Astapenko","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Artem","family":"Kruchinkin","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Novik","family":"Vladimir","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kirill","family":"Arzamasov","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2025,7,15]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"118904","DOI":"10.1016\/j.eswa.2022.118904","article-title":"An efficient method for autoencoder based outlier detection","volume":"213","author":"Abhaya","year":"2023","journal-title":"Expert Syst. Appl."},{"key":"ref2","doi-asserted-by":"publisher","first-page":"708948","DOI":"10.1155\/2015\/708948","article-title":"Data transformation technique to improve the outlier detection power of Grubbs\u2019 test for data expected to follow linear relation","volume":"2015","author":"Adikaram","year":"2015","journal-title":"J. Appl. Math."},{"key":"ref3","doi-asserted-by":"publisher","first-page":"891","DOI":"10.1093\/biomet\/74.4.891","article-title":"Masking effect on tests for outliers in normal samples","volume":"74","author":"Bendre","year":"1987","journal-title":"Biometrika"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"113","DOI":"10.1007\/s12018-009-9064-4","article-title":"T-scores and Z-scores","volume":"8","author":"Carey","year":"2010","journal-title":"Clin. Rev. Bone Miner. Metab."},{"key":"ref5","doi-asserted-by":"publisher","first-page":"782","DOI":"10.1080\/01621459.1993.10476339","article-title":"The identification of multiple outliers","volume":"88","author":"Davies","year":"1993","journal-title":"J. Am. Stat. Assoc."},{"key":"ref6","doi-asserted-by":"publisher","first-page":"39","DOI":"10.14714\/CP80.1314","article-title":"Understanding the heat map","volume":"80","author":"DeBoer","year":"2015","journal-title":"Cartogr. Perspect."},{"key":"ref7","volume-title":"\u201cOn the Development of Artificial Intelligence in the Russian Federation\u201d No. 490","year":"2019"},{"key":"ref8","first-page":"15","article-title":"International conference on image analysis and processing","volume-title":"ICIAP 2022: Image analysis and processing \u2013 ICIAP 2022; 2022 may 23\u201327","author":"Diers","year":"2022"},{"key":"ref9","first-page":"1","article-title":"An overview of machine learning-based techniques for detecting outliers in data","volume":"6","author":"Dulesov","year":"2023","journal-title":"Eng. J. Don"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"297","DOI":"10.1007\/s41060-021-00265-1","article-title":"On the nature and types of anomalies: a review of deviations in data","volume":"12","author":"Foorthuis","year":"2021","journal-title":"Int. J. Data Sci. Anal."},{"key":"ref11","volume-title":"Splenomegaly. Reference article, Radiopaedia.org","author":"Gaillard","year":"2009"},{"key":"ref12","author":"Gaspar","year":"2011"},{"key":"ref13","author":"Goldstein","year":"2012"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"100","DOI":"10.1038\/s43586-022-00184-w","article-title":"Principal component analysis","volume":"2","author":"Greenacre","year":"2022","journal-title":"Nat. Rev. Methods Primers"},{"key":"ref15","author":"Ji","year":"2017"},{"key":"ref16","author":"Kim","year":"2019"},{"key":"ref17","first-page":"70","article-title":"Overview of data cleaning methods for machine learning","volume":"11","author":"Makarov","year":"2023","journal-title":"Int. J. Open Inf. Technol."},{"key":"ref18","doi-asserted-by":"publisher","first-page":"1050","DOI":"10.6026\/973206300181050","article-title":"Current trends on the application of artificial intelligence in medical sciences","volume":"18","author":"Mashraqi","year":"2022","journal-title":"Bioinformation"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"110","DOI":"10.12928\/telkomnika.v17i1.9394","article-title":"Analysis of DBSCAN and K-means algorithm for evaluating outlier on RFM model of customer behaviour","volume":"17","author":"Monalisa","year":"2019","journal-title":"Telecommun. Comput. Electron. Control"},{"key":"ref20","volume-title":"Experiment on the use of innovative computer vision technologies for medical image analysis and subsequent applicability in the healthcare system of Moscow","year":"2025"},{"key":"ref21","doi-asserted-by":"crossref","first-page":"152","DOI":"10.55145\/ajest.2023.02.02.018","article-title":"Detecting data outliers with machine learning","volume":"2","author":"Nassreddine","year":"2023","journal-title":"Al-Salam J. Eng. Technol."},{"key":"ref22","doi-asserted-by":"publisher","first-page":"161","DOI":"10.1086\/100259","article-title":"Criterion for the rejection of doubtful observations","volume":"2","author":"Peirce","year":"1852","journal-title":"Astron. J."},{"key":"ref23","first-page":"1460","article-title":"Anomaly detection in the dataset using unsupervised learning algorithms isolation Forest and local outlier factor","volume":"12","author":"Popova","year":"2020","journal-title":"StudNet"},{"key":"ref24","year":"2025"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"3829","DOI":"10.17576\/jsm-2022-5111-26","article-title":"A comparison of efficiency of test statistics for detecting outliers in normal population","volume":"51","author":"Promtep","year":"2022","journal-title":"Sains Malaysiana"},{"key":"ref26","first-page":"799","volume-title":"Metrology, standardization and certification","author":"Radkevich","year":"2006"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"221","DOI":"10.2307\/1268354","article-title":"On the detection of many outliers","volume":"17","author":"Rosner","year":"1975","journal-title":"Technometrics"},{"key":"ref28","doi-asserted-by":"publisher","first-page":"165","DOI":"10.1080\/00401706.1983.10487848","article-title":"Percentage points for a generalized ESD many-outlier procedure","volume":"25","author":"Rosner","year":"1983","journal-title":"Technometrics"},{"key":"ref29","doi-asserted-by":"publisher","first-page":"19","DOI":"10.1145\/3068335","article-title":"DBSCAN revisited, revisited: why and how you should (still) use DBSCAN","volume":"42","author":"Schubert","year":"2017","journal-title":"ACM Trans. Database Syst."},{"key":"ref30","doi-asserted-by":"publisher","first-page":"4","DOI":"10.21683\/1729-2646-2024-24-1-4-9","article-title":"Methodology for detecting and removing outliers in statistical studies","volume":"24","author":"Sidnyaev","year":"2024","journal-title":"Dependability"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"642","DOI":"10.1198\/016214504000001466","article-title":"Outlier labeling with boxplot procedures","volume":"100","author":"Sim","year":"2005","journal-title":"J. Am. Stat. Assoc."},{"key":"ref32","author":"Sysoev","year":"2019"},{"key":"ref33","volume-title":"Certificate of state registration of a computer program no. 2024680469, Russian Federation. Software module for loading, selecting, and de-identifying studies in DICOM format stored in the unified radiological information system of Moscow: No. 2024667717","author":"Vasilev","year":"2025"},{"key":"ref34","volume-title":"Preparing datasets for training and testing of artificial intelligence-powered software","author":"Vasilev","year":"2024"},{"key":"ref35","doi-asserted-by":"publisher","first-page":"28","DOI":"10.21045\/1811-0185-2023-4-28-41","article-title":"Medical datasets for machine learning: fundamental principles of standartization and systematization","volume":"4","author":"Vasilev","year":"2023","journal-title":"Manag. Zdravookhran."},{"key":"ref36","doi-asserted-by":"publisher","first-page":"93","DOI":"10.17816\/DD321423","article-title":"Double-reading mammograms using artificial intelligence technologies: a new model of mass preventive examination organization","volume":"4","author":"Vasilev","year":"2023","journal-title":"Digit. Diagn."},{"key":"ref37","volume-title":"Dataset curation platform 2025","author":"Vasilev","year":"2025"},{"key":"ref38","first-page":"511","article-title":"Detection of outliers using interquartile range technique from intrusion dataset","volume-title":"Information and decision sciences. Advances in intelligent systems and computing","author":"Vinutha","year":"2018"},{"key":"ref39","doi-asserted-by":"publisher","first-page":"669","DOI":"10.1007\/s42081-020-00091-y","article-title":"Outliers in official statistics","volume":"3","author":"Wada","year":"2020","journal-title":"Jpn. J. Stat. Data Sci."},{"key":"ref40","author":"Wang","year":"2021"},{"key":"ref41","doi-asserted-by":"publisher","first-page":"198","DOI":"10.1016\/j.patcog.2017.09.012","article-title":"Hyperparameter selection of one-class support vector machine by self-adaptive data shifting","volume":"74","author":"Wang","year":"2018","journal-title":"Pattern Recogn."},{"key":"ref42","doi-asserted-by":"publisher","first-page":"174830262210781","DOI":"10.1177\/17483026221078111","article-title":"Outlier detection algorithm based on k-nearest neighbors-local outlier factor","volume":"16","author":"Xu","year":"2022","journal-title":"J. Algorithms Comput. Technol."},{"key":"ref43","doi-asserted-by":"publisher","first-page":"101946","DOI":"10.1016\/j.datak.2021.101946","article-title":"Anomaly explanation: a review","volume":"137","author":"Yepmo","year":"2022","journal-title":"Data Knowl. Eng."},{"key":"ref44","doi-asserted-by":"publisher","first-page":"335","DOI":"10.1037\/a0020140","article-title":"Fitting data to model: structural equation modeling diagnosis using two scatter plots","volume":"15","author":"Yuan","year":"2010","journal-title":"Psychol. Methods"},{"key":"ref45","first-page":"16","article-title":"About single outliers tests","volume":"7","author":"Zalyazhnyh","year":"2022","journal-title":"J. Nat. Sci. Res."},{"key":"ref46","volume-title":"Context-encoding variational autoencoder for unsupervised anomaly detection","author":"Zimmerer","year":"2022"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1607348\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,15]],"date-time":"2025-07-15T05:25:58Z","timestamp":1752557158000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1607348\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,15]]},"references-count":46,"alternative-id":["10.3389\/frai.2025.1607348"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1607348","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,15]]},"article-number":"1607348"}}