{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,31]],"date-time":"2026-10-31T20:37:32Z","timestamp":1793479052703,"version":"4.5.0"},"reference-count":34,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,2,19]],"date-time":"2026-02-19T00:00:00Z","timestamp":1771459200000},"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. Public Health"],"abstract":"<jats:p>Advances in diffusion-based and neural rendering architectures have enabled the creation of synthetic audiovisual content that closely replicates natural facial dynamics, speech production, and environmental context. These developments pose a growing risk to clinical medicine and dentistry, where authentic audiovisual data support remote clinical assessment, communication, and medico-legal documentation. This study introduces an interpretable multimodal framework for deepfake detection that integrates visual, acoustic, and cross-modal coherence features, with decision thresholds derived exclusively from authentic recordings to ensure transparency and forensic accountability. Using the DeepFake RealWorld dataset comprising 46,371 audiovisual clips, including 77% with audio, we evaluated 47 descriptors across optical, bioacoustic, and synchronization domains. Clinical relevance was evaluated through simulated dental teleconsultations. Cross-modal metrics, particularly lip-speech synchrony (\u0394p\u202f=\u202f0.21\u20130.22), phoneme-viseme alignment (\u0394p\u202f=\u202f0.21), a widely used audio visual consistency cue in multimodal deepfake detection, identity coherence (\u0394p\u202f=\u202f0.19), and scene-audio semantic consistency (\u0394p\u202f=\u202f0.18) demonstrated the strongest discriminatory performance, with prevalence ratios of up to 2.7. Acoustic markers, including reduced jitter, shimmer, and shortened reverberation time (RT60; 0.12\u202fs in synthetic vs. 0.28\u202fs in real recordings), provided additional robustness. The framework maintained performance degradation below 15% under platform-scale compression and recapture artifacts. Additionally, the proposed framework was benchmarked against a standard open-source texture-oriented baseline detector based on the Xception architecture, with clip-level ROC AUC and balanced accuracy reported on the original clips and under the same platform transformations used in the robustness analysis. Simulated dental teleconsultations revealed that manipulated recordings introduce inconsistencies in mandibular motion, prosody-related facial dynamics, and ambient acoustic plausibility (mean \u0394p\u202f=\u202f0.18; PR\u202f=\u202f2.3), confirming the clinical relevance of multimodal coherence analysis. These results position coherence-based detection as a reliable, transparent, and domain-appropriate approach for safeguarding audiovisual integrity in remote dentistry, medicine, and related digital health applications.<\/jats:p>","DOI":"10.3389\/fpubh.2026.1781216","type":"journal-article","created":{"date-parts":[[2026,2,19]],"date-time":"2026-02-19T06:42:57Z","timestamp":1771483377000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Generative AI-driven synthetic media risks in digital health: implications for telemedicine and teledentistry"],"prefix":"10.3389","volume":"14","author":[{"given":"Karol","family":"J\u0119drasiak","sequence":"first","affiliation":[{"name":"WSB University","place":["Dabrowa Gornicza, Poland"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Julia","family":"Bijoch","sequence":"additional","affiliation":[{"name":"Collegium Medicum - Faculty of Medicine, WSB University","place":["Dabrowa Gornicza, Poland"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,2,19]]},"reference":[{"key":"ref1","author":"Pei","year":"2024"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"910","DOI":"10.1109\/jstsp.2020.3002101","article-title":"Media forensics and deepfakes: an overview","volume":"14","author":"Verdoliva","year":"2020","journal-title":"IEEE J Sel Top Signal Process"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"791","DOI":"10.1007\/s40122-024-00620-7","article-title":"Innovative applications of telemedicine and other digital health solutions in pain management: a literature review","volume":"13","author":"El-Tallawy","year":"2024","journal-title":"Pain Ther"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2025.3586963","article-title":"Securing healthcare data integrity: deepfake detection using autonomous AI approaches","author":"Hsu","year":"2025","journal-title":"IEEE J Biomed Health Inform"},{"key":"ref5","volume-title":"Beyond deception: exploiting deepfake technology for ethical innovation in healthcare","author":"Kaur","year":"2024"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.38124\/ijisrt\/25oct811","article-title":"Securing the digital health ecosystem: a framework for safeguarding patient data in telemedicine","author":"Berko","year":"2025","journal-title":"Int J Innov Sci Res Technol"},{"key":"ref7","doi-asserted-by":"crossref","DOI":"10.1109\/ICCIT65724.2025.11167460","article-title":"Digital health innovations and telemedicine effectiveness in remote patient monitoring","author":"Susilawati","year":"2025"},{"key":"ref8","author":"Hashmi","year":"2024"},{"key":"ref9","first-page":"1","article-title":"FaceForensics++: learning to detect manipulated facial images","author":"Rossler","year":"2019"},{"key":"ref10","volume-title":"Multimodaltrace: deepfake detection using audiovisual representation learning","author":"Raza","year":"2023"},{"key":"ref11","first-page":"1596","article-title":"PIA: Deepfake detection using phoneme-temporal and identity-dynamic analysis","author":"Datta","year":"2025"},{"key":"ref12","author":"Shahzad","year":"2023"},{"key":"ref13","author":"Du","year":"2025"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"1","DOI":"10.55041\/IJSREM52704","article-title":"Multimodal deepfake detection frameworks: a survey","volume":"9","author":"Inchara","year":"2025","journal-title":"Int J Sci Res Eng Manag"},{"key":"ref15","author":"Tariq","year":"2025"},{"key":"ref16","article-title":"Integration of multimodal coherence features for robust and explainable deepfake detection with clinical dentistry use cases","author":"J\u0119drasiak","year":"2025"},{"key":"ref17","author":"Dolhansky","year":"2020"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1007\/s12652-024-04856-1","article-title":"Explainable biometrics: a systematic literature review","author":"Tucci","year":"2024","journal-title":"J Ambient Intell Human Comput"},{"key":"ref19","volume-title":"Recommendation R128: Loudness normalisation and permitted maximum level of audio signals","year":"2011"},{"key":"ref20","volume-title":"BS.1770\u20134: Algorithms to measure audio programme loudness and true-peak audio level","year":"2015"},{"key":"ref21","author":"Wang","year":"2025"},{"key":"ref22","doi-asserted-by":"crossref","DOI":"10.4324\/9780080499253","volume-title":"Applied photographic optics","author":"Ray","year":"2002"},{"key":"ref23","first-page":"402","article-title":"RAFT: recurrent all-pairs field transforms for optical flow","author":"Teed","year":"2020"},{"key":"ref24","volume-title":"3382\u20131: Acoustics \u2013 Measurement of room acoustic parameters - part 1: Performance spaces","year":"2009"},{"key":"ref25","first-page":"7884","article-title":"Adverb: visually guided audio dereverberation","author":"Chowdhury","year":"2023"},{"key":"ref26","first-page":"166","article-title":"HiFi-GAN-2: studio-quality speech enhancement via generative adversarial networks conditioned on acoustic features","author":"Su","year":"2021"},{"key":"ref27","volume-title":"BT.1359\u20131: Relative timing of sound and vision for broadcasting","year":"1998"},{"key":"ref28","first-page":"4690","article-title":"ArcFace: additive angular margin loss for deep face recognition","author":"Deng","year":"2019"},{"key":"ref29","author":"Desplanques","year":"2020"},{"key":"ref30","first-page":"112","article-title":"Multimodal speech emotion recognition using audio and text","author":"Yoon","year":"2018"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"289","DOI":"10.1111\/j.2517-6161.1995.tb02031.x","article-title":"Controlling the false discovery rate: a practical and powerful approach to multiple testing","volume":"57","author":"Benjamini","year":"1995","journal-title":"J R Stat Soc Series B Stat Methodol"},{"key":"ref32","first-page":"1251","article-title":"Xception: deep learning with depthwise separable convolutions","author":"Chollet","year":"2017"},{"key":"ref33","author":"Doshi-Velez","year":"2017"},{"key":"ref34","first-page":"7890","article-title":"Watch your up-convolution: CNN-based generative deep neural networks are failing to reproduce spectral distributions","author":"Durall","year":"2020"}],"container-title":["Frontiers in Public Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2026.1781216\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,19]],"date-time":"2026-02-19T06:42:58Z","timestamp":1771483378000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2026.1781216\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,19]]},"references-count":34,"alternative-id":["10.3389\/fpubh.2026.1781216"],"URL":"https:\/\/doi.org\/10.3389\/fpubh.2026.1781216","relation":{},"ISSN":["2296-2565"],"issn-type":[{"value":"2296-2565","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,19]]},"article-number":"1781216"}}