{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T18:22:41Z","timestamp":1791483761691,"version":"4.3.4"},"reference-count":32,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T00:00:00Z","timestamp":1778457600000},"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. Digit. Health"],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Cervical cancer remains a leading cause of cancer mortality in resource-constrained settings, where access to advanced digital pathology equipment is severely limited. Automated histopathological image segmentation offers a potential pathway to improve diagnostic access, but practical solutions combining affordable hardware with robust deep learning remain underdeveloped.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We present an approach combining smartphone-assisted microscopy with DeepLabV3+ architecture for precise segmentation of cervical cancer lesions in H&amp;amp;E-stained histopathological images. A custom smartphone adapter and the Ocular data collection app were used for standardized image acquisition. A DeepLabV3+ model with ResNet34 encoder was developed and validated on 5,966 histopathological images collected from the Uganda Cancer Institute, targeting 21 distinct histopathological feature classes using a combined BCE and Dice loss function with memory-efficient training on an NVIDIA RTX 3090.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>On a held-out validation set drawn from the same institutional dataset, the system achieves a mean Intersection over Union (IoU) of 75.8% and a Dice coefficient of 93.1%, leveraging atrous spatial pyramid pooling to capture multi-scale contextual information. Per-class IoU ranged narrowly from 74.13% to 75.41% across all 21 feature classes, demonstrating consistent segmentation performance. DeepLabV3+ outperformed a U-Net baseline trained under identical conditions (mIoU: 56.84%, Dice: 68.53%), confirming the architectural contribution of the pre-trained encoder and ASPP module.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>These results demonstrate the technical feasibility of reliable digital pathology analysis in resource-limited settings using readily available smartphone hardware. The DeepLabV3+ architecture's superior boundary delineation and multi-scale feature extraction prove particularly effective for complex histopathological patterns. These results are reported on validation data only; independent multi-institutional evaluation will be necessary to assess generalization to broader clinical populations and imaging conditions before any clinical deployment.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fdgth.2026.1776155","type":"journal-article","created":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T05:30:14Z","timestamp":1778477414000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Adapting DeepLabV3+ for biopsy cervical cancer lesion segmentation"],"prefix":"10.3389","volume":"8","author":[{"given":"Rose","family":"Nakasi","sequence":"first","affiliation":[{"name":"Artificial Intelligence Health Lab, Department of Computer Science, Makerere University","place":["Kampala, Uganda"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Cosmas","family":"Wamozo","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Makerere University","place":["Kampala, Uganda"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Solomon","family":"Nsumba","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Makerere University","place":["Kampala, Uganda"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Benjamin","family":"Rukundo","sequence":"additional","affiliation":[{"name":"Artificial Intelligence Health Lab, Department of Computer Science, Makerere University","place":["Kampala, Uganda"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tonny","family":"Okecha","sequence":"additional","affiliation":[{"name":"Department of Pathology, Uganda Cancer Institute","place":["Kampala, Uganda"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Byron","family":"Mubiru","sequence":"additional","affiliation":[{"name":"Artificial Intelligence Health Lab, Department of Computer Science, Makerere University","place":["Kampala, Uganda"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chodrine","family":"Mutebi","sequence":"additional","affiliation":[{"name":"Artificial Intelligence Health Lab, Department of Computer Science, Makerere University","place":["Kampala, Uganda"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,5,11]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"e197","DOI":"10.1016\/S2214-109X(22)00501-0","article-title":"Global estimates of incidence and mortality of cervical cancer in 2020: a baseline analysis of the who global cervical cancer elimination initiative","volume":"11","author":"Singh","year":"2023","journal-title":"Lancet Glob Health"},{"key":"B2","doi-asserted-by":"publisher","first-page":"72","DOI":"10.1200\/EDBK_280625","article-title":"Cancer prevention in low-resource countries: an overview of the opportunity","volume":"40","author":"Kamaraju","year":"2020","journal-title":"Am Soc Clin Oncol Educ Book"},{"key":"B3","doi-asserted-by":"publisher","first-page":"170","DOI":"10.35335\/laweco.v17i3.43","article-title":"Impact of poverty reduction programs on healthcare access in remote Ar-eas: fostering community development for sustainable health","volume":"17","author":"Amri","year":"2023","journal-title":"Law Econ"},{"key":"B4","doi-asserted-by":"publisher","first-page":"254","DOI":"10.1038\/s41377-024-01597-w","article-title":"Towards next-generation diagnostic pathology: AI-empowered label-free multiphoton microscopy","volume":"13","author":"Wang","year":"2024","journal-title":"Light Sci Appl"},{"key":"B5","doi-asserted-by":"publisher","first-page":"1940","DOI":"10.3390\/healthcare10101940","article-title":"Recent advancements in emerging technologies for healthcare management systems: a survey","volume":"10","author":"Junaid","year":"2022","journal-title":"Healthcare"},{"key":"B6","doi-asserted-by":"publisher","first-page":"2636","DOI":"10.1364\/BOE.511384","article-title":"Deep learning-assisted smartphone-based quantitative microscopy for label-free peripheral blood smear analysis","volume":"15","author":"Huang","year":"2024","journal-title":"Biomed Opt Express"},{"key":"B7","doi-asserted-by":"publisher","first-page":"e0009677","DOI":"10.1371\/journal.pntd.0009677","article-title":"Mobile microscopy and telemedicine platform assisted by deep learning for the quantification of trichuris trichiura infection","volume":"15","author":"Dacal","year":"2021","journal-title":"PLoS Negl Trop Dis"},{"key":"B8","author":"Chen","year":""},{"key":"B9","doi-asserted-by":"publisher","first-page":"686","DOI":"10.1200\/GO.21.00013","article-title":"Feasibility and acceptability of smartphone-based cervical cancer screening among HIV-positive women in Western Kenya","volume":"7","author":"Mungo","year":"2021","journal-title":"JCO Glob Oncol"},{"key":"B10","doi-asserted-by":"publisher","first-page":"455","DOI":"10.2147\/IJWH.S136351","article-title":"Cervical cancer screening in low-resource settings: a smartphone image application as an alternative to colposcopy","volume":"9","author":"Gallay","year":"2017","journal-title":"Int J Womens Health"},{"key":"B11","doi-asserted-by":"publisher","first-page":"151","DOI":"10.1007\/s12551-023-01175-1","article-title":"A commentary on the development and use of smartphone imaging devices","volume":"16","author":"Raju","year":"2023","journal-title":"Biophys Rev"},{"key":"B12","doi-asserted-by":"publisher","first-page":"6047","DOI":"10.3390\/cancers13236047","article-title":"Use of smartphones for the detection of uterine cervical cancer: a systematic review","volume":"13","author":"Champin","year":"2021","journal-title":"Cancers"},{"key":"B13","author":"Ronneberger","year":""},{"key":"B14","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1016\/j.media.2019.02.010","article-title":"Medical image classification using synergic deep learning","volume":"54","author":"Zhang","year":"2019","journal-title":"Med Image Anal."},{"key":"B15","doi-asserted-by":"publisher","first-page":"2372","DOI":"10.3390\/cancers14102372","article-title":"Fully automatic whole-volume tumor segmentation in cervical cancer","volume":"14","author":"Hodneland","year":"2022","journal-title":"Cancers"},{"key":"B16","doi-asserted-by":"publisher","first-page":"1297","DOI":"10.1007\/s00330-019-06467-3","article-title":"Deep learning for fully automated tumor segmentation and extraction of magnetic resonance radiomics features in cervical cancer","volume":"30","author":"Lin","year":"2020","journal-title":"Eur Radiol"},{"key":"B17","doi-asserted-by":"publisher","first-page":"195","DOI":"10.3390\/computation11100195","article-title":"A robust deep learning approach for accurate segmentation of cytoplasm and nucleus in noisy pap smear images","volume":"11","author":"Nazir","year":"2023","journal-title":"Computation"},{"key":"B18","doi-asserted-by":"publisher","first-page":"101897","DOI":"10.1016\/j.artmed.2020.101897","article-title":"A shape context fully convolutional neural network for segmentation and classification of cervical nuclei in pap smear images","volume":"107","author":"Hussain","year":"2020","journal-title":"Artif Intell Med"},{"key":"B19","doi-asserted-by":"publisher","first-page":"3968","DOI":"10.1002\/mp.14898","article-title":"Automatic clinical target volume delineation for cervical cancer in CT images using deep learning","volume":"48","author":"Shi","year":"2021","journal-title":"Med Phys"},{"key":"B20","doi-asserted-by":"publisher","first-page":"13650","DOI":"10.1038\/s41598-022-18084-0","article-title":"Evaluation of auto-segmentation for EBRT planning structures using deep learning-based workflow on cervical cancer","volume":"12","author":"Wang","year":"2022","journal-title":"Sci Rep"},{"key":"B21","doi-asserted-by":"publisher","first-page":"10","DOI":"10.4103\/jpi.jpi_53_19","article-title":"EpithNet: deep regression for epithelium segmentation in cervical histology images","volume":"11","author":"Sornapudi","year":"2020","journal-title":"J Pathol Inform"},{"key":"B22","doi-asserted-by":"publisher","first-page":"983","DOI":"10.1007\/s10278-022-00620-z","article-title":"The accuracy and radiomics feature effects of multiple U-Net-based automatic segmentation models for transvaginal ultrasound images of cervical cancer","volume":"35","author":"Jin","year":"2022","journal-title":"J Digit Imaging"},{"key":"B23","doi-asserted-by":"publisher","first-page":"5176","DOI":"10.21037\/qims-24-369","article-title":"Cervical cancer segmentation based on medical images: a literature review","volume":"14","author":"Wang","year":"2024","journal-title":"Quant Imaging Med Surg"},{"key":"B24","doi-asserted-by":"publisher","first-page":"103520","DOI":"10.1016\/j.micron.2023.103520","article-title":"Recent developments in cervical cancer diagnosis using deep learning on whole slide images: An overview of models, techniques, challenges and future directions","volume":"173","author":"Sambyal","year":"2023","journal-title":"Micron"},{"key":"B25","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1016\/j.media.2017.07.005","article-title":"A survey on deep learning in medical image analysis","volume":"42","author":"Litjens","year":"2017","journal-title":"Med Image Anal"},{"key":"B26","doi-asserted-by":"publisher","first-page":"9375","DOI":"10.1109\/ACCESS.2017.2788044","article-title":"Deep learning applications in medical image analysis","volume":"6","author":"Ker","year":"2017","journal-title":"IEEE Access"},{"key":"B27","author":"Tan","year":""},{"key":"B28","author":"Kornblith","year":""},{"key":"B29","author":"Raghu","year":""},{"key":"B30","first-page":"1","article-title":"Metrics for evaluating 3D medical image segmentation: analysis, selection, and tool","volume":"15","author":"Aziz Taha","year":"2015","journal-title":"BMC Med Imaging"},{"key":"B31","author":"Reinke","year":""},{"key":"B32","first-page":"125","article-title":"Albumentations: fast and flexible image augmentations","volume-title":"Information","author":"Buslaev","year":"2020"}],"container-title":["Frontiers in Digital Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1776155\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T05:30:15Z","timestamp":1778477415000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1776155\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,11]]},"references-count":32,"alternative-id":["10.3389\/fdgth.2026.1776155"],"URL":"https:\/\/doi.org\/10.3389\/fdgth.2026.1776155","relation":{},"ISSN":["2673-253X"],"issn-type":[{"value":"2673-253X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,11]]},"article-number":"1776155"}}