Explainable Deep Learning Framework for Periodontal Diagnosis in Dental Imaging

preprint OA: closed
Full text JSON View at publisher
AI-generated deep summary by claude@2026-07, 2026-07-03 · read from full text

This preprint studied an automated, explainable deep learning framework to classify periapical dental radiographs into periodontal versus normal categories, using a convolutional neural network combined with explainable AI via Grad-CAM. The authors describe training a CNN architecture with four convolutional layers (ReLU and max-pooling) and two fully connected layers, applying data augmentation and optimizing with stochastic gradient descent with momentum, reporting 94.17% classification accuracy with stable convergence and Grad-CAM heatmaps that highlighted relevant regions, such as alveolar bone loss in periodontal cases. A major caveat explicitly stated is that the work is a preprint and has not been peer reviewed. This paper does not explicitly discuss endometriosis or adenomyosis; it was included in the corpus via a keyword match in the upstream search index.

Read from the paper's body, not the abstract. Not a substitute for reading the paper. No clinical advice. How this works

Abstract

Abstract Periodontal disease is a chronic inflammatory condition impacting tooth-supporting structures. Early detection via radiographic analysis is essential to prevent progression and tooth loss. However, radiograph interpretation remains subjective and reliant on clinical expertise. This study presents an automated diagnostic framework using a Convolutional Neural Network (CNN) combined with Explainable Artificial Intelligence (XAI) through Gradient-weighted Class Activation Mapping (Grad-CAM). The system classifies periapical dental radiographs into periodontal and normal categories. The proposed CNN includes four convolutional layers with ReLU activation, max-pooling, and two fully connected layers. Data augmentation improved generalizability, and training employed the Stochastic Gradient Descent with Momentum (SGDM) algorithm. The model achieved a 94.17% classification accuracy with stable convergence. Grad-CAM heatmaps highlighted relevant regions influencing the model's decisions. In periodontal cases, alveolar bone loss was consistently identified, while normal images showed no pathological patterns. This visualization enhances interpretability and clinical trust. The integration of CNN and Grad-CAM offers a reliable, transparent diagnostic tool for dental radiology and supports the ethical adoption of AI in healthcare.
Full text 26,845 characters · extracted from preprint-html · click to expand
Explainable Deep Learning Framework for Periodontal Diagnosis in Dental Imaging | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Explainable Deep Learning Framework for Periodontal Diagnosis in Dental Imaging Yusra Fadhillah, Muhammad Noor Hasan Siregar, Ade Ismail Abdul Kodir, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7458110/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Periodontal disease is a chronic inflammatory condition impacting tooth-supporting structures. Early detection via radiographic analysis is essential to prevent progression and tooth loss. However, radiograph interpretation remains subjective and reliant on clinical expertise. This study presents an automated diagnostic framework using a Convolutional Neural Network (CNN) combined with Explainable Artificial Intelligence (XAI) through Gradient-weighted Class Activation Mapping (Grad-CAM). The system classifies periapical dental radiographs into periodontal and normal categories. The proposed CNN includes four convolutional layers with ReLU activation, max-pooling, and two fully connected layers. Data augmentation improved generalizability, and training employed the Stochastic Gradient Descent with Momentum (SGDM) algorithm. The model achieved a 94.17% classification accuracy with stable convergence. Grad-CAM heatmaps highlighted relevant regions influencing the model's decisions. In periodontal cases, alveolar bone loss was consistently identified, while normal images showed no pathological patterns. This visualization enhances interpretability and clinical trust. The integration of CNN and Grad-CAM offers a reliable, transparent diagnostic tool for dental radiology and supports the ethical adoption of AI in healthcare. Periodontal diagnosis Deep learning Explainable AI Dental imaging Medical image analysis Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7458110","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":522813849,"identity":"9125038e-5d15-4623-81f7-4bca2605f869","order_by":0,"name":"Yusra Fadhillah","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCElEQVRIie3PMUvDQBTA8XfcenSXSOIXEF44cDHWb+F8oZBbUsjYMaXQSZwtGfwKLYLzwQ1dBFehiyI4dThwEczgSydRrnEUvP92D368dwCh0N8tBrC85oBZ3D3NUz+RRBiRqpA7on5FTEeczevuvY/go35174DJYM2mbxUW+ubCPtOWYXxc+0gpF5eA6cKyWXSN2Xi1KZDISJ4YL+EgoGVL+ksksBivGtURk995iX5hLeA5kdmHQKvTRrseoiQXgDmROW2xKonK/VsO7reSHyKO6C/zUzosXUZlZRT6/zJY02HbCZ5dPVi7EW2WJI2+dW4yjH3kaDfHr6f+mHwrqfsnoVAo9N/7BOx4Y8HphKLvAAAAAElFTkSuQmCC","orcid":"","institution":"Graha Nusantara University","correspondingAuthor":true,"prefix":"","firstName":"Yusra","middleName":"","lastName":"Fadhillah","suffix":""},{"id":522813850,"identity":"0a9229d6-ee0b-45f1-bfbc-990264cb08a8","order_by":1,"name":"Muhammad Noor Hasan Siregar","email":"","orcid":"","institution":"Graha Nusantara University","correspondingAuthor":false,"prefix":"","firstName":"Muhammad","middleName":"Noor Hasan","lastName":"Siregar","suffix":""},{"id":522813852,"identity":"9dbb624d-198c-423e-aa6b-7935f766aecc","order_by":2,"name":"Ade Ismail Abdul Kodir","email":"","orcid":"","institution":"Sultan Agung Islamic University","correspondingAuthor":false,"prefix":"","firstName":"Ade","middleName":"Ismail Abdul","lastName":"Kodir","suffix":""},{"id":522813858,"identity":"01813aba-98dc-4fa5-990a-4be09d062051","order_by":3,"name":"Khairur Rizki","email":"","orcid":"","institution":"Andalas University","correspondingAuthor":false,"prefix":"","firstName":"Khairur","middleName":"","lastName":"Rizki","suffix":""}],"badges":[],"createdAt":"2025-08-26 03:08:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7458110/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7458110/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":92603098,"identity":"3e742fbb-aff1-4a51-94b5-b2f8b763ce03","added_by":"auto","created_at":"2025-10-01 14:54:49","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1003762,"visible":true,"origin":"","legend":"","description":"","filename":"ExplainableDeepLearningFrameworkforPeriodontalDiagnosisinDentalImagingRV1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/7ed9cf1132cf364a4c1199cd.docx"},{"id":92603181,"identity":"d9aabd3e-72f4-45a3-8c3b-69864f807aef","added_by":"auto","created_at":"2025-10-01 14:55:31","extension":"jpg","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":162251,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1a.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/b3d8a7770032ac7f497c8360.jpg"},{"id":92603123,"identity":"6aa90b69-9537-4282-bd20-212e68ac3d87","added_by":"auto","created_at":"2025-10-01 14:55:21","extension":"jpg","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":121224,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1b.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/d90bcfb81543e0c34fcba74b.jpg"},{"id":92603035,"identity":"baad17d0-4ce4-455e-b809-3d2464520312","added_by":"auto","created_at":"2025-10-01 14:54:39","extension":"jpg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":52484,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/58b8f8b9ca95c4a3671d0f45.jpg"},{"id":92603132,"identity":"265dcd8d-6dc5-4862-a5fa-a95f1f825ced","added_by":"auto","created_at":"2025-10-01 14:55:23","extension":"jpg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":52284,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/1ca22b537c45cb9c3c4043cb.jpg"},{"id":92603121,"identity":"bd6f43c1-fd08-4b76-ab31-2b8dae0d7972","added_by":"auto","created_at":"2025-10-01 14:55:21","extension":"jpg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29196,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4a.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/05a03b10f6c687c147f47a04.jpg"},{"id":92603280,"identity":"cf9809ab-5c7a-4679-8a24-e8f3ed25f360","added_by":"auto","created_at":"2025-10-01 14:56:04","extension":"jpg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37034,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4b.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/2ca0c8b4e42fb57bcf331969.jpg"},{"id":92603096,"identity":"ac9350be-3c5d-44fa-bad2-71b213434c0a","added_by":"auto","created_at":"2025-10-01 14:54:48","extension":"json","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6431,"visible":true,"origin":"","legend":"","description":"","filename":"f8fe0b7364324b0897c5fb0e86ef6aba.json","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/9f5dbfb5507e594276c7e640.json"},{"id":92603126,"identity":"634bc87a-433f-4dc7-8323-9aa5ef03fe1a","added_by":"auto","created_at":"2025-10-01 14:55:22","extension":"xml","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":97573,"visible":true,"origin":"","legend":"","description":"","filename":"f8fe0b7364324b0897c5fb0e86ef6aba1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/bd0f9ca0fc31a8ae2aabd8fd.xml"},{"id":92603118,"identity":"06d97d18-27bb-4b1c-ae21-2aba632f25ab","added_by":"auto","created_at":"2025-10-01 14:55:19","extension":"jpg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":162251,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1a.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/5a8564b6a50a9742af28842d.jpg"},{"id":92603164,"identity":"76fb521a-3048-47a1-afad-75481fa3e2d2","added_by":"auto","created_at":"2025-10-01 14:55:27","extension":"jpg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":121224,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1b.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/305f0a9237525fbb068c6c70.jpg"},{"id":92603220,"identity":"d438cd66-a2c9-4c02-be4d-3853b72ce3e9","added_by":"auto","created_at":"2025-10-01 14:55:51","extension":"jpg","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":52484,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/4545a10ca980626cb6827d97.jpg"},{"id":92603009,"identity":"4e421575-8cc2-4957-ae82-6e68971f20b0","added_by":"auto","created_at":"2025-10-01 14:54:37","extension":"jpg","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":52284,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/334e9b8624ef21cff5af5fd5.jpg"},{"id":92603184,"identity":"c96ebb3b-a266-4dfa-9455-712f904ba424","added_by":"auto","created_at":"2025-10-01 14:55:33","extension":"jpg","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29196,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4a.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/d05fac718cf5022f6ef80fc5.jpg"},{"id":92603188,"identity":"4732bbe6-0b5e-4346-8cbf-a881836060f9","added_by":"auto","created_at":"2025-10-01 14:55:34","extension":"jpg","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37034,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4b.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/5761bfdf7837e45aa406cfff.jpg"},{"id":92603133,"identity":"cadc0bb1-5846-4eb2-bbf2-c774e8d36e64","added_by":"auto","created_at":"2025-10-01 14:55:23","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1399,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/2e1243e3726b78df5b93836b.png"},{"id":92603033,"identity":"7c401969-8373-44a3-88c0-5863d6f2a63d","added_by":"auto","created_at":"2025-10-01 14:54:38","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1399,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/842b8834f5174bd88e8937f6.png"},{"id":92603206,"identity":"ccdb6f0d-b783-444d-952b-91ae492605e9","added_by":"auto","created_at":"2025-10-01 14:55:45","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1399,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/28668c68fea33d6f0d0f3a9a.png"},{"id":92603122,"identity":"0b61c279-eddc-4f63-9687-075b37d654fa","added_by":"auto","created_at":"2025-10-01 14:55:21","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1399,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/26b82d04f51bec4c199864d7.png"},{"id":92603224,"identity":"b719831a-9de5-4456-baf1-dbb256d1a1eb","added_by":"auto","created_at":"2025-10-01 14:55:53","extension":"jpeg","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":748307,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/4be396c7a8319cdd72218520.jpeg"},{"id":92603171,"identity":"eaa37432-7118-4047-9ad3-85a61ba6ca35","added_by":"auto","created_at":"2025-10-01 14:55:28","extension":"jpeg","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":232416,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/911fd7fb211e677ed47c9ab4.jpeg"},{"id":92603099,"identity":"adf21780-9775-42ad-b6bf-9a51b88b6340","added_by":"auto","created_at":"2025-10-01 14:54:52","extension":"jpeg","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":236784,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/26c7dc481a2ead69b36d4e35.jpeg"},{"id":92603114,"identity":"596e28fb-d52b-4e0e-b9f3-cb40c2274427","added_by":"auto","created_at":"2025-10-01 14:55:06","extension":"jpeg","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":219045,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/71343074dfde9b0f90add473.jpeg"},{"id":92603120,"identity":"f7a3c75e-70e4-4f0a-8ecb-088c089f47ea","added_by":"auto","created_at":"2025-10-01 14:55:19","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":406659,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure1a.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/b2cab6222287d99033e7d89a.png"},{"id":92603180,"identity":"4f74c6db-5385-46cb-91a0-4212ac4a68a1","added_by":"auto","created_at":"2025-10-01 14:55:30","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":47783,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure1b.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/711f1f51bbd04d735740afeb.png"},{"id":92602983,"identity":"972942aa-70e1-4fff-a168-e1a6511ead45","added_by":"auto","created_at":"2025-10-01 14:54:28","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11081,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/7253fc080c5d47ee24552c7d.png"},{"id":92603161,"identity":"72be205e-67f5-498b-9f3f-59e1a56d0263","added_by":"auto","created_at":"2025-10-01 14:55:26","extension":"png","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11520,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/014b137e8ddd9636a2b8557a.png"},{"id":92603183,"identity":"ff4a7106-e7be-4327-a103-e927cefe2215","added_by":"auto","created_at":"2025-10-01 14:55:31","extension":"png","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8023,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure4a.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/617c96397022e481923d829a.png"},{"id":92603254,"identity":"92c362cb-ef6a-41c4-94a5-50e2f9bba39a","added_by":"auto","created_at":"2025-10-01 14:55:59","extension":"png","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10311,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure4b.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/cf4f6cb270e437eef1fc19e6.png"},{"id":92603113,"identity":"0f3a0548-17c4-4914-beee-ad7c4812e24e","added_by":"auto","created_at":"2025-10-01 14:55:04","extension":"png","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1162,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/718bda7df7e326f14107189b.png"},{"id":92603115,"identity":"512d8683-83bd-4c5d-a947-66d86ed5f235","added_by":"auto","created_at":"2025-10-01 14:55:07","extension":"png","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1162,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/3c335906f71901e596e630f2.png"},{"id":92603185,"identity":"95c528c9-3587-4373-ba04-a51066f496a6","added_by":"auto","created_at":"2025-10-01 14:55:33","extension":"png","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1162,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/bca2b8f3bf2c6680fb2da09c.png"},{"id":92603187,"identity":"c3ddc022-fff2-46e6-b64a-56e237ea3aae","added_by":"auto","created_at":"2025-10-01 14:55:33","extension":"png","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1162,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/12c08673806071fb3b04d6c3.png"},{"id":92603250,"identity":"04b93e67-52d4-49b5-b4f0-e2dc060e6d57","added_by":"auto","created_at":"2025-10-01 14:55:57","extension":"png","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1103235,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/07c65c2c4ef7baa462168cc5.png"},{"id":92603124,"identity":"df26726c-3059-4dae-9d75-5ec43d295e60","added_by":"auto","created_at":"2025-10-01 14:55:22","extension":"png","order_by":34,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":58651,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/48b174bab0e378b54a4b9c31.png"},{"id":92603029,"identity":"8b092123-f17c-4bcb-b25d-76ac1e757684","added_by":"auto","created_at":"2025-10-01 14:54:38","extension":"png","order_by":35,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":60420,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/eabaf3f2cdeab8d3bb47b405.png"},{"id":92603182,"identity":"0ba8db86-a67f-420d-95b8-ec22264fac4a","added_by":"auto","created_at":"2025-10-01 14:55:31","extension":"png","order_by":36,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56379,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/b3f178da1a17adc6416cba1e.png"},{"id":92603119,"identity":"ab9fd8e4-9685-4ca0-b953-1a712b400040","added_by":"auto","created_at":"2025-10-01 14:55:19","extension":"xml","order_by":37,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":96614,"visible":true,"origin":"","legend":"","description":"","filename":"f8fe0b7364324b0897c5fb0e86ef6aba1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/31b663b7bd502b7439ea2938.xml"},{"id":92603034,"identity":"ab4a8c7e-a243-4738-886f-91ce1e2f2542","added_by":"auto","created_at":"2025-10-01 14:54:39","extension":"html","order_by":38,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":109386,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1/34fbc2704793351537a06b86.html"},{"id":93942116,"identity":"61dda31d-571c-439e-a960-221c235a0ccc","added_by":"auto","created_at":"2025-10-20 13:47:12","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":622205,"visible":true,"origin":"","legend":"","description":"","filename":"ExplainableDeepLearningFrameworkforPeriodontalDiagnosisinDentalImagingRV1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7458110/v1_covered_bb3716a2-f285-4869-9d4a-ea340d7266ba.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Explainable Deep Learning Framework for Periodontal Diagnosis in Dental Imaging","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Periodontal diagnosis, Deep learning, Explainable AI, Dental imaging, Medical image analysis","lastPublishedDoi":"10.21203/rs.3.rs-7458110/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7458110/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Periodontal disease is a chronic inflammatory condition impacting tooth-supporting structures. Early detection via radiographic analysis is essential to prevent progression and tooth loss. However, radiograph interpretation remains subjective and reliant on clinical expertise. This study presents an automated diagnostic framework using a Convolutional Neural Network (CNN) combined with Explainable Artificial Intelligence (XAI) through Gradient-weighted Class Activation Mapping (Grad-CAM). The system classifies periapical dental radiographs into periodontal and normal categories. The proposed CNN includes four convolutional layers with ReLU activation, max-pooling, and two fully connected layers. Data augmentation improved generalizability, and training employed the Stochastic Gradient Descent with Momentum (SGDM) algorithm. The model achieved a 94.17% classification accuracy with stable convergence. Grad-CAM heatmaps highlighted relevant regions influencing the model's decisions. In periodontal cases, alveolar bone loss was consistently identified, while normal images showed no pathological patterns. This visualization enhances interpretability and clinical trust. The integration of CNN and Grad-CAM offers a reliable, transparent diagnostic tool for dental radiology and supports the ethical adoption of AI in healthcare.","manuscriptTitle":"Explainable Deep Learning Framework for Periodontal Diagnosis in Dental Imaging","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-01 14:24:35","doi":"10.21203/rs.3.rs-7458110/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"9aea8ce9-075f-4d90-b34b-98fe947fb5b5","owner":[],"postedDate":"October 1st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-10-20T13:38:49+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-01 14:24:35","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7458110","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7458110","identity":"rs-7458110","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}

Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy (via DOI) is the canonical version.

My notes (saved in your browser only)

Ask this paper AI returns verbatim quotes from the full text · source: preprint-html

Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works

Citation neighborhood (no data yet)

We don't have any in-corpus citations linked to this paper yet. This is a recent paper (2025) — citers typically take a year or two to land, and the OpenAlex reference graph may still be filling in.

Source provenance

europepmc
last seen: 2026-05-20T01:45:00.602351+00:00