{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T06:19:29Z","timestamp":1787552369382,"version":"build-2736575974"},"reference-count":33,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T00:00:00Z","timestamp":1787529600000},"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. Immunol."],"abstract":"\n Introduction<\/jats:title>\n \n Acute promyelocytic leukemia (APL) is defined by its genetics, yet interphase fluorescence\n in situ<\/jats:italic>\n hybridization (FISH) for the PML::RARA fusion, the chromosomal complement seen on a metaphase spread, and the marrow\u2019s cellular composition on a Wright\u2013Giemsa smear are read as separate, unconnected examinations. We did not identify a published model trained on two or more of these image types in the same patients with APL.\n <\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n We describe two image-derived readouts for a retrospective cohort of patients with APL. A genetic complexity axis classifies the FISH fusion signal pattern and flags any additional cytogenetic abnormality (ACA) beyond the t(15;17), directly from raw, unarranged metaphase spreads. A marrow composition axis extends a cytomorphology pipeline, built on frozen DinoBloom embeddings and a per-cell classification head directly supervised on an annotated subset, from promyelocyte detection to a seven-class differential count. From this composition, we compute a residual hematopoiesis index, the model-estimated fraction of non-leukemic nucleated marrow cells, so that a higher value indicates a larger non-leukemic fraction. The index is a morphology-derived surrogate for residual non-leukemic marrow composition and carries no immune-phenotypic information. The association between the two axes is computed directly from patient-level, out-of-fold branch outputs; a small, three-node fusion module over the two genetic and one composition readout supports only a secondary, exploratory analysis, reported in the Supplementary Material. The cohort held 700 patients with smear, FISH, and karyotype linked per patient, of whom 650 had complete clinical annotation; we evaluated each branch with patient-stratified fivefold cross-validation, five repeats, pooled out-of-fold, and report bootstrap confidence intervals (CIs) over patients. Diagnostic flow cytometry was retrievable for 118 patients, and 80 slides were re-scanned with fields independently re-selected, to check the index against an external measurement and against itself.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n The fusion-pattern classifier had a macro-averaged F1 of 0.82 across the five signal classes, and the karyotype ACA reader had an area under the curve (AUC) of 0.86. The residual hematopoiesis index had a median of 0.42 (interquartile range, IQR, 0.29 to 0.58) across the cohort, tracked the flow-derived non-blast fraction at a Spearman correlation of 0.80 (95% CI 0.72 to 0.86) in the 118 patients with flow data, and had an intraclass correlation of 0.91 on re-scanned slides. Genetic complexity and the residual hematopoiesis index were associated at a Spearman\n \u03c1<\/jats:italic>\n of 0.24 (95% CI 0.17 to 0.31,\n p &lt;<\/jats:italic>\n 0.0001), the primary finding of this study, with higher genetic complexity accompanying a larger non-leukemic fraction. The association held under adjustment for metaphase yield, scorable nuclei per FISH image, age, presenting white-cell count, and accession year, with partial correlations between 0.19 and 0.24, and patients with an expert-defined additional abnormality had a higher index than those with the isolated translocation (median 0.46 against 0.41). An exploratory test of the same readouts against the Sanz risk group and DS occurrence, run on the clinically annotated subcohort, showed performance no better than chance, and an adjusted logistic model bounded any effect of an additional cytogenetic abnormality on DS at an odds ratio of 1.07 (95% CI 0.74 to 1.55), in line with two prior clinical cohorts that reported no association between cytogenetic abnormality and DS.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusion<\/jats:title>\n Image-derived genetic complexity was positively associated with the residual hematopoiesis index, and the association persisted under the adjustments and substitutions applied to it. These findings describe a cross-sectional association in a single-center, retrospective cohort, not a validated diagnostic or risk tool. The residual hematopoiesis index summarizes morphology alone, its agreement with flow cytometry is supportive rather than confirmatory, and the secondary DS analysis drew on a smaller, clinically annotated subset of the cohort rather than on every patient.<\/jats:p>\n <\/jats:sec>","DOI":"10.3389\/fimmu.2026.1933105","type":"journal-article","created":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T05:30:50Z","timestamp":1787549450000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Multimodal AI reading of genetic complexity and residual marrow composition in acute promyelocytic leukemia"],"prefix":"10.3389","volume":"17","author":[{"given":"Minjie","family":"Gao","sequence":"first","affiliation":[{"name":"Department of Hematology, The First Affiliated Hospital of Ningbo University","place":["Ningbo, Zhejiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guifang","family":"Ouyang","sequence":"additional","affiliation":[{"name":"Department of Hematology, The First Affiliated Hospital of Ningbo University","place":["Ningbo, Zhejiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yangguang","family":"Liu","sequence":"additional","affiliation":[{"name":"College of Artificial Intelligence, Ningbo University of Finance and Economics","place":["Ningbo, Zhejiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shikun","family":"Chen","sequence":"additional","affiliation":[{"name":"College of Artificial Intelligence, Ningbo University of Finance and Economics","place":["Ningbo, Zhejiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,24]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1182\/blood-2013-10-512640","article-title":"How we prevent and treat differentiation syndrome in patients with acute promyelocytic leukemia","volume":"123","author":"Sanz","year":"2014","journal-title":"Blood"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.3389\/fimmu.2018.01357","article-title":"Arsenic trioxide enhances the NK cell cytotoxicity against acute promyelocytic leukemia while simultaneously inhibiting its bio-genesis","volume":"9","author":"Abu","year":"2018","journal-title":"Front Immunol"},{"key":"B3","doi-asserted-by":"publisher","first-page":"261","DOI":"10.1186\/s13059-025-03736-x","article-title":"U-FISH: a fluorescent spot detector for imaging-based spatial-omics analysis and AI-assisted FISH diagnosis","volume":"26","author":"Xu","year":"2025","journal-title":"Genome Biol"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1544","DOI":"10.1182\/blood-2024-211644","article-title":"Karyotype AI for Precision Oncology","volume":"144","author":"Fang","year":"2024","journal-title":"Blood"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.1038\/s41698-021-00179-y","article-title":"Deep learning for diagnosis of acute promyelocytic leukemia via recognition of genomically imprinted morphologic features","volume":"5","author":"Sidhom","year":"2021","journal-title":"NPJ Precis Oncol"},{"key":"B6","doi-asserted-by":"publisher","first-page":"201","DOI":"10.1186\/s12885-022-09307-8","article-title":"Deep learning identifies acute promyelocytic leukemia in bone marrow smears","volume":"22","author":"Eckardt","year":"2022","journal-title":"BMC Cancer"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.3390\/jcm14051670","article-title":"Applications of artificial intelligence in acute promyelocytic leukemia: An avenue of opportunities? a systematic review","volume":"14","author":"G\u0103man","year":"2025","journal-title":"J Clin Med"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1097\/md.0000000000046458","article-title":"Analysis of risk factors for differentiation syndrome in patients with acute promyelocytic leukemia","volume":"104","author":"Zhang","year":"2025","journal-title":"Medicine"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.1002\/cam4.6398","article-title":"Additional cytogenetic abnormalities in patients with newly diagnosed acute promyelocytic leukemia predict inferior event-free survival","volume":"12","author":"Zeng","year":"2023","journal-title":"Cancer Med"},{"key":"B10","doi-asserted-by":"publisher","first-page":"2562","DOI":"10.1038\/s41598-023-29160-4","article-title":"Detection of acute promyelocytic leukemia in peripheral blood and bone marrow with annotation-free deep learning","volume":"13","author":"Manescu","year":"2023","journal-title":"Sci Rep"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.1182\/bloodadvances.2023010975","article-title":"Deep learning predicts therapy-relevant genetics in acute myeloid leukemia from pappenheim-stained bone marrow smears","volume":"8","author":"Kockwelp","year":"2024","journal-title":"Blood Adv"},{"key":"B12","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-72390-2_49","article-title":"DinoBloom: A foundation model for generalizable cell embeddings in hematology","volume-title":"Medical Image Computing and Computer Assisted Intervention (MICCAI) 2024","author":"Koch","year":"2024"},{"key":"B13","doi-asserted-by":"publisher","first-page":"8231","DOI":"10.1038\/s41598-019-44643-z","article-title":"Automated detection of the HER2 gene amplification status in fluorescence in situ hybridization images for the diagnostics of cancer tissues","volume":"9","author":"Zakrzewski","year":"2019","journal-title":"Sci Rep"},{"key":"B14","doi-asserted-by":"publisher","first-page":"9746","DOI":"10.1038\/s41598-023-36811-z","article-title":"Deep learning to automatically evaluate HER2 gene amplification status from fluorescence in situ hybridization images","volume":"13","author":"Xue","year":"2023","journal-title":"Sci Rep"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2019.2905841","article-title":"Varifocal-Net: A chromosome classification approach using deep convolutional networks","volume":"38","author":"Qin","year":"2019","journal-title":"IEEE Trans Med Imaging"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.1109\/tmi.2020.3007642","article-title":"DeepACEv2: Automated chromosome enumeration in metaphase cell images using deep convolutional neural networks","volume":"39","author":"Xiao","year":"2020","journal-title":"IEEE Trans Med Imaging"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1007\/s11684-024-1098-y","article-title":"ChromTR: chromosome detection in raw metaphase cell images via deformable transformers","volume":"18","author":"Xia","year":"2024","journal-title":"Front Med"},{"key":"B18","article-title":"Multimodal prototyping for cancer survival prediction","volume-title":"Proceedings of the 41st International Conference on Machine Learning (ICML) 2024","author":"Song","year":"2024"},{"key":"B19","article-title":"Prototypical information bottlenecking and disentangling for multimodal cancer survival prediction","volume-title":"International Conference on Learning Representations (ICLR) 2024","author":"Zhang","year":"2024"},{"key":"B20","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR52733.2024.01099","article-title":"Morphological prototyping for unsupervised slide representation learning in computational pathology","volume-title":"IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2024","author":"Song","year":"2024"},{"key":"B21","doi-asserted-by":"crossref","DOI":"10.52202\/079017-2057","article-title":"HEALNet: Multimodal fusion for heterogeneous biomedical data","volume-title":"Advances in Neural Information Processing Systems (Neurips) 2024","author":"Hemker","year":"2024"},{"key":"B22","article-title":"Gradient-guided modality decoupling for missingmodality robustness","volume-title":"Proceedings of the AAAI Conference on Artificial Intelligence","author":"Wang","year":"2024"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.1016\/j.ebiom.2025.105663","article-title":"Deep learning informed multimodal fusion of radiology and pathology to predict outcomes in HPV-associated oropharyngeal squamous cell carcinoma","volume":"114","author":"Song","year":"2025","journal-title":"eBioMedicine"},{"key":"B24","article-title":"Multi-stain self-attention graph multiple instance learning pipeline for histopathology whole slide images","volume-title":"British Machine Vision Conference (Bmvc) 2023","author":"Gallagher-Syed","year":"2023"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.1182\/blood-2008-07-168617","article-title":"Differentiation syndrome in patients with acute promyelocytic leukemia treated with all-trans retinoic acid and anthracycline chemotherapy: characteristics, outcome, and prognostic factors","volume":"113","author":"Montesinos","year":"2009","journal-title":"Blood"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.1182\/bloodadvances.2021006682","article-title":"Effect of additional cytogenetic abnormalities on survival in arsenic trioxide-treated acute promyelocytic leukemia","volume":"6","author":"Epstein-Peterson","year":"2022","journal-title":"Blood Adv"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-020-01018-x","article-title":"Cellpose: a generalist algorithm for cellular segmentation","volume":"18","author":"Stringer","year":"2021","journal-title":"Nat Methods"},{"key":"B28","article-title":"DINOv2: Learning robust visual features without supervision","author":"Oquab","year":"2024","journal-title":"Trans Mach Learn Res"},{"key":"B29","doi-asserted-by":"publisher","first-page":"187","DOI":"10.1038\/s41698-026-01383-4","article-title":"A comprehensive foundation model for generalizable cytogenetics in precision oncology with CHROMA","volume":"10","author":"Yang","year":"2026","journal-title":"NPJ Precis Oncol"},{"key":"B30","doi-asserted-by":"publisher","DOI":"10.1136\/bmj-2023-078378","article-title":"TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods","volume":"385","author":"Collins","year":"2024","journal-title":"BMJ"},{"key":"B31","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.2020200029","article-title":"Checklist for artificial intelligence in medical imaging (CLAIM): A guide for authors and reviewers","volume":"2","author":"Mongan","year":"2020","journal-title":"Radiology: Artif Intell"},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.240300","article-title":"Checklist for artificial intelligence in medical imaging (CLAIM): 2024 update","volume":"6","author":"Tejani","year":"2024","journal-title":"Radiology: Artif Intell"},{"key":"B33","doi-asserted-by":"crossref","DOI":"10.18653\/v1\/2020.acl-main.385","article-title":"Quantifying attention flow in transformers","volume-title":"Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL 2020)","author":"Abnar","year":"2020"}],"container-title":["Frontiers in Immunology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fimmu.2026.1933105\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T05:30:53Z","timestamp":1787549453000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fimmu.2026.1933105\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,24]]},"references-count":33,"alternative-id":["10.3389\/fimmu.2026.1933105"],"URL":"https:\/\/doi.org\/10.3389\/fimmu.2026.1933105","relation":{},"ISSN":["1664-3224"],"issn-type":[{"value":"1664-3224","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,24]]},"article-number":"1933105"}}