{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,12]],"date-time":"2026-01-12T11:12:02Z","timestamp":1768216322766,"version":"3.49.0"},"reference-count":40,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,1,12]],"date-time":"2026-01-12T00:00:00Z","timestamp":1768176000000},"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>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Radiotherapy (RT) remains a cornerstone treatment for head and neck cancer squamous cell carcinoma. However, therapeutic responses vary considerably among patients due to radiation resistance, which limits long-term survival and contributes to recurrence and disease progression. Developing robust deep learning (DL) and machine learning (ML)-based predictive models is essential to improve response prediction, evaluate treatment outcomes, and identify biomarkers linked to radiosensitization.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      This single-center retrospective study applied DL and ML models to analyze CT scans and RNA-seq gene expression data for prognostic and biomarker discovery purposes. For image analyses, two independent datasets were used. Dataset A includes 1,100 CT scans (pre- and post-treatment) from 476 patients with stage III and IV laryngeal carcinoma treated with response-adapted RT. A convolutional neural network (CNNs) integrated with a recurrent network (RNNs) was used for single-point tumor localization and response prediction. Dataset B, comprising 500 scans from 169 patients treated with radical RT, served as the additional validation cohort. Pre- and post-treatment scans were used to train a DL model, which showed better prediction performance for survival and disease-specific outcomes, including progression and locoregional recurrence. For gene expression-based biomarker analysis, TCGA data (\n                      <jats:italic>n<\/jats:italic>\n                      \u202f=\u202f231) were examined using glmBoost, support vector machine classifier (SVM), and random forest (RF) algorithms to construct and predict genes associated with radiosensitivity, and the GSE20020 dataset was used to validate the model performance. Proteins and mRNA were used to confirm the signature biomarkers using qRT-PCR and LC\u2013MS mass spectrometry.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Findings<\/jats:title>\n                    <jats:p>\n                      For CT scan image analysis, the DL-model achieved AUCs of 0.792 (\n                      <jats:italic>p<\/jats:italic>\n                      \u202f=\u202f0.031) at 2-month and 0.832 (\n                      <jats:italic>p<\/jats:italic>\n                      \u202f&amp;lt;\u202f0.01) at 6-month follow-up. Risk scores significantly correlated with overall survival (HR 1.59, 95% CI 1.34\u20133.22,\n                      <jats:italic>p<\/jats:italic>\n                      \u202f=\u202f0.063), progression-free survival (1.39, 95% CI 1.16\u20132.29,\n                      <jats:italic>p<\/jats:italic>\n                      \u202f=\u202f0.103). The pathological response in dataset B was likewise significantly predicted by the model. Among 39 differentially expressed genes, ML-model analysis identified 13 candidate genes associated with radiosensitivity on repeated cross-validation with an AUROC of 0.91 in the training set. In the validation dataset, when the models were optimized, the models consistently predicted seven core genes, achieving AUCs ranging from 0.96 to 0.94 to predict the radiosensitivity.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Interpretation<\/jats:title>\n                    <jats:p>These findings highlight the effectiveness of DL and ML approaches in integrating imaging and transcriptomic data to predict response-adapted RT response and patient outcomes. These automated, and interpretable AI-driven biomarkers hold significant potential for clinical translation. Future research should aim to expand datasets and validate the models in multicenter cohorts for broader applicability.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/frai.2025.1738174","type":"journal-article","created":{"date-parts":[[2026,1,12]],"date-time":"2026-01-12T08:11:07Z","timestamp":1768205467000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Deep learning and machine learning integration of radiomics and transcriptomics predicts response-adapted radiotherapy outcome and radiosensitivity in resectable locally advanced laryngeal carcinoma"],"prefix":"10.3389","volume":"8","author":[{"given":"Shafat","family":"Ujjahan","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Abu Shadat M.","family":"Noman","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sarah S.","family":"Al-Johani","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zakia","family":"Shinwari","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ayodele A.","family":"Alaiya","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Syed S.","family":"Islam","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,1,12]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"121541","DOI":"10.1016\/J.LFS.2023.121541","article-title":"Targeting aldehyde dehydrogenase enzymes in combination with chemotherapy and immunotherapy: an approach to tackle resistance in cancer cells","volume":"320","author":"Al-Shamma","year":"2023","journal-title":"Life Sci."},{"key":"ref2","doi-asserted-by":"publisher","first-page":"4313","DOI":"10.2147\/JIR.S322430","article-title":"Alterations in the plasma proteome induced by SARS-CoV-2 and MERS-CoV reveal biomarkers for disease outcomes for COVID-19 patients","volume":"14","author":"Alaiya","year":"2021","journal-title":"J. Inflamm. Res."},{"key":"ref3","doi-asserted-by":"publisher","first-page":"104938","DOI":"10.1016\/j.ebiom.2023.104938","article-title":"Machine learning-driven identification of the gene-expression signature associated with a persistent multiple organ dysfunction trajectory in critical illness","volume":"99","author":"Atreya","year":"2024","journal-title":"EBioMedicine"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"592303","DOI":"10.3389\/FIMMU.2021.592303\/FULL","article-title":"Machine learning identifies complicated sepsis course and subsequent mortality based on 20 genes in peripheral blood immune cells at 24 H post-ICU admission","volume":"12","author":"Banerjee","year":"2021","journal-title":"Front. Immunol."},{"key":"ref5","doi-asserted-by":"publisher","first-page":"100590","DOI":"10.1016\/J.CTRO.2023.100590","article-title":"Artificial intelligence to predict outcomes of head and neck radiotherapy","volume":"39","author":"Bang","year":"2023","journal-title":"Clin. Transl. Radiat. Oncol."},{"key":"ref6","doi-asserted-by":"publisher","first-page":"3948","DOI":"10.1158\/1078-0432.CCR-20-4935","article-title":"Deep learning for fully automated prediction of overall survival in patients with oropharyngeal cancer using FDG-PET imaging","volume":"27","author":"Cheng","year":"2021","journal-title":"Clin. Cancer Res."},{"key":"ref7","doi-asserted-by":"publisher","first-page":"1088","DOI":"10.1002\/IJC.21954","article-title":"KIF14 mRNA expression is a predictor of grade and outcome in breast cancer","volume":"119","author":"Corson","year":"2006","journal-title":"Int. J. Cancer"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"442","DOI":"10.1016\/J.JSS.2014.01.017","article-title":"The survival predictive significance of HOXC6 and HOXC8 in esophageal squamous cell carcinoma","volume":"188","author":"Du","year":"2014","journal-title":"J. Surg. Res."},{"key":"ref9","doi-asserted-by":"publisher","first-page":"594","DOI":"10.1016\/J.IJROBP.2015.03.004","article-title":"Total laryngectomy versus larynx preservation for T4a larynx cancer: patterns of care and survival outcomes","volume":"92","author":"Grover","year":"2015","journal-title":"Int. J. Radiat. Oncol. Biol. Phys."},{"key":"ref10","doi-asserted-by":"publisher","first-page":"1744","DOI":"10.3174\/AJNR.A2177","article-title":"Treatment response assessment of head and neck cancers on CT using computerized volume analysis","volume":"31","author":"Hadjiiski","year":"2010","journal-title":"AJNR Am. J. Neuroradiol."},{"key":"ref11","doi-asserted-by":"publisher","first-page":"18148","DOI":"10.3390\/IJMS151018148","article-title":"TPX2 is a prognostic marker and contributes to growth and metastasis of human hepatocellular carcinoma","volume":"15","author":"Huang","year":"2014","journal-title":"Int. J. Mol. Sci."},{"key":"ref12","first-page":"448","article-title":"Batch normalization: accelerating deep network training by reducing internal covariate shift","author":"Ioffe","year":"2015"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/S41419-024-06512-0","article-title":"Ovarian tumor cell-derived JAGGED2 promotes omental metastasis through stimulating the notch signaling pathway in the mesothelial cells","volume":"15","author":"Islam","year":"2024","journal-title":"Cell Death Dis."},{"key":"ref14","doi-asserted-by":"publisher","first-page":"669","DOI":"10.1016\/J.BIOPHA.2013.04.005","article-title":"The role of aldehyde dehydrogenase (ALDH) in cancer drug resistance","volume":"67","author":"Januchowski","year":"2013","journal-title":"Biomed. Pharmacother."},{"key":"ref15","doi-asserted-by":"publisher","first-page":"32","DOI":"10.1038\/S12276-022-00907-9","article-title":"TPX2 prompts mitotic survival via the induction of BCL2L1 through YAP1 protein stabilization in human embryonic stem cells","volume":"55","author":"Kim","year":"2023","journal-title":"Exp. Mol. Med."},{"key":"ref16","doi-asserted-by":"publisher","first-page":"84","DOI":"10.1145\/3065386","article-title":"ImageNet classification with deep convolutional neural networks","volume":"60","author":"Krizhevsky","year":"2017","journal-title":"Commun. ACM"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"447","DOI":"10.1016\/J.CRITREVONC.2013.05.005","article-title":"Larynx preservation: what is the best non-surgical strategy?","volume":"88","author":"Lagha","year":"2013","journal-title":"Crit. Rev. Oncol. Hematol."},{"key":"ref18","doi-asserted-by":"publisher","first-page":"4648","DOI":"10.1097\/JS9.0000000000001578","article-title":"MRI-based deep learning and radiomics for prediction of occult cervical lymph node metastasis and prognosis in early-stage oral and oropharyngeal squamous cell carcinoma: a diagnostic study","volume":"110","author":"Lan","year":"2024","journal-title":"Int. J. Surg."},{"key":"ref19","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41598-022-07034-5","article-title":"Cross-institutional outcome prediction for head and neck cancer patients using self-attention neural networks","volume":"12","author":"Le","year":"2022","journal-title":"Sci. Rep."},{"key":"ref20","doi-asserted-by":"publisher","first-page":"2708","DOI":"10.1093\/ANNONC\/MDS065","article-title":"Laryngeal preservation with induction chemotherapy for hypopharyngeal squamous cell carcinoma: 10-year results of EORTC trial 24891","volume":"23","author":"Lefebvre","year":"2012","journal-title":"Ann. Oncol."},{"key":"ref21","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1155\/2022\/8656865","article-title":"Expression profiles of HOXC6 predict the survival of glioblastoma patients and correlate with cell cycle","volume":"2022","author":"Li","year":"2022","journal-title":"J. Oncol."},{"key":"ref22","doi-asserted-by":"publisher","first-page":"1279","DOI":"10.1038\/S41598-025-85498-X","article-title":"Identification of CT based radiomic biomarkers for progression free survival in head and neck squamous cell carcinoma","volume":"15","author":"Ling","year":"2025","journal-title":"Sci. Rep."},{"key":"ref23","doi-asserted-by":"publisher","first-page":"103","DOI":"10.1001\/JAMAOTO.2022.3996","article-title":"Association of primary tumor volume with survival in patients with T3 glottic cancer treated with radiotherapy: a study of the Canadian Head & Neck Collaborative Research Initiative","volume":"149","author":"Malik","year":"2023","journal-title":"JAMA Otolaryngol. Head Neck Surg."},{"key":"ref24","doi-asserted-by":"publisher","first-page":"101841","DOI":"10.1016\/J.TRANON.2023.101841","article-title":"Prognostic value of hypoxia-responsive gene expression profile in patients diagnosed with head and neck squamous cell carcinoma","volume":"39","author":"Matic","year":"2023","journal-title":"Transl. Oncol."},{"key":"ref25","doi-asserted-by":"publisher","first-page":"855","DOI":"10.1001\/JAMAOTO.2014.1671","article-title":"Survival outcomes in advanced laryngeal cancer","volume":"140","author":"Megwalu","year":"2014","journal-title":"JAMA Otolaryngol. Head Neck Surgery"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"35678","DOI":"10.1074\/JBC.M112.361675","article-title":"HOXC6 is deregulated in human head and neck squamous cell carcinoma and modulates Bcl-2 expression","volume":"287","author":"Moon","year":"2012","journal-title":"J. Biol. Chem."},{"key":"ref27","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/S41525-022-00293-1","article-title":"Precision drugging of the MAPK pathway in head and neck cancer","volume":"7","author":"Ngan","year":"2022","journal-title":"NPJ Genom. Med."},{"key":"ref28","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/S13046-022-02370-W","article-title":"Stanniocalcin 2 (STC2): a universal tumour biomarker and a potential therapeutical target","volume":"41","author":"Qie","year":"2022","journal-title":"J. Exp. Clin. Cancer Res."},{"key":"ref29","doi-asserted-by":"publisher","first-page":"109","DOI":"10.1001\/JAMAONCOL.2024.5356","article-title":"Deep learning model for predicting immunotherapy response in advanced non-small cell lung Cancer","volume":"11","author":"Rakaee","year":"2025","journal-title":"JAMA Oncol."},{"key":"ref30","first-page":"1929","article-title":"Dropout: a simple way to prevent neural networks from overfitting","volume":"15","author":"Srivastava","year":"2014","journal-title":"J. Mach. Learn. Res."},{"key":"ref31","doi-asserted-by":"publisher","first-page":"345","DOI":"10.1007\/S12105-014-0597-6\/METRICS","article-title":"Gene expression characterization of HPV positive head and neck cancer to predict response to chemoradiation","volume":"9","author":"Thibodeau","year":"2015","journal-title":"Head Neck Pathol."},{"key":"ref32","first-page":"139","volume-title":"Glottic cancer. Clinical cases in ENT","author":"Williamson","year":"2023"},{"key":"ref33","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/S12935-019-0833-Y","article-title":"Identification of ESM1 overexpressed in head and neck squamous cell carcinoma","volume":"19","author":"Xu","year":"2019","journal-title":"Cancer Cell Int."},{"key":"ref34","doi-asserted-by":"publisher","first-page":"3266","DOI":"10.1158\/1078-0432.CCR-18-2495","article-title":"Deep learning predicts lung cancer treatment response from serial medical imaging","volume":"25","author":"Xu","year":"2019","journal-title":"Clin. Cancer Res."},{"key":"ref35","doi-asserted-by":"publisher","first-page":"4514","DOI":"10.1016\/J.RADCR.2023.09.058","article-title":"Complete response of glottic cancer to intra-arterial infusion chemotherapy combined with radiotherapy: a report of 4 cases","volume":"18","author":"Yamakuni","year":"2023","journal-title":"Radiol Case Rep"},{"key":"ref36","doi-asserted-by":"publisher","first-page":"5976","DOI":"10.18632\/ONCOTARGET.13355","article-title":"STC2 promotes head and neck squamous cell carcinoma metastasis through modulating the PI3K\/AKT\/snail signaling","volume":"8","author":"Yang","year":"2016","journal-title":"Oncotarget"},{"key":"ref37","doi-asserted-by":"publisher","first-page":"44842","DOI":"10.18632\/ONCOTARGET.15107","article-title":"Phase III randomized trial of preoperative concurrent chemoradiotherapy versus preoperative radiotherapy for patients with locally advanced head and neck squamous cell carcinoma","volume":"8","author":"Yi","year":"2017","journal-title":"Oncotarget"},{"key":"ref38","doi-asserted-by":"publisher","first-page":"178","DOI":"10.1016\/J.ORALONCOLOGY.2019.06.020","article-title":"The prognostic value of CT-based image-biomarkers for head and neck cancer patients treated with definitive (chemo-)radiation","volume":"95","author":"Zhai","year":"2019","journal-title":"Oral Oncol."},{"key":"ref39","doi-asserted-by":"publisher","first-page":"439","DOI":"10.4149\/NEO_2013_057","article-title":"Upregulated Hoxc6 expression is associated with poor survival in gastric cancer patients","volume":"60","author":"Zhang","year":"2013","journal-title":"Neoplasma"},{"key":"ref40","doi-asserted-by":"publisher","first-page":"2563","DOI":"10.18632\/AGING.205498","article-title":"Integrative analysis confirms TPX2 as a novel biomarker for clinical implication, tumor microenvironment, and immunotherapy response across human solid tumors","volume":"16","author":"Zhu","year":"2024","journal-title":"Aging"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1738174\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,12]],"date-time":"2026-01-12T08:11:10Z","timestamp":1768205470000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1738174\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,12]]},"references-count":40,"alternative-id":["10.3389\/frai.2025.1738174"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1738174","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,12]]},"article-number":"1738174"}}