A Multimodal Deep Learning Approach to Predicting Systemic Diseases from Oral Conditions
<b>Background:</b> It is known that oral diseases such as periodontal (gum) disease are closely linked to various systemic diseases and disorders. Deep learning advances have the potential to make major contributions to healthcare, particularly in the domains that rely on medical imaging...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/12/12/3192 |
| _version_ | 1797460575948636160 |
|---|---|
| author | Dan Zhao Morteza Homayounfar Zhe Zhen Mei-Zhen Wu Shuk Yin Yu Kai-Hang Yiu Varut Vardhanabhuti George Pelekos Lijian Jin Mohamad Koohi-Moghadam |
| author_facet | Dan Zhao Morteza Homayounfar Zhe Zhen Mei-Zhen Wu Shuk Yin Yu Kai-Hang Yiu Varut Vardhanabhuti George Pelekos Lijian Jin Mohamad Koohi-Moghadam |
| author_sort | Dan Zhao |
| collection | DOAJ |
| description | <b>Background:</b> It is known that oral diseases such as periodontal (gum) disease are closely linked to various systemic diseases and disorders. Deep learning advances have the potential to make major contributions to healthcare, particularly in the domains that rely on medical imaging. Incorporating non-imaging information based on clinical and laboratory data may allow clinicians to make more comprehensive and accurate decisions. <b>Methods:</b> Here, we developed a multimodal deep learning method to predict systemic diseases and disorders from oral health conditions. A dual-loss autoencoder was used in the first phase to extract periodontal disease-related features from 1188 panoramic radiographs. Then, in the second phase, we fused the image features with the demographic data and clinical information taken from electronic health records (EHR) to predict systemic diseases. We used receiver operation characteristics (ROC) and accuracy to evaluate our model. The model was further validated by an unseen test dataset. <b>Findings:</b> According to our findings, the top three most accurately predicted chapters, in order, are the Chapters III, VI and IX. The results indicated that the proposed model could predict systemic diseases belonging to Chapters III, VI and IX, with AUC values of 0.92 (95% CI, 0.90–94), 0.87 (95% CI, 0.84–89) and 0.78 (95% CI, 0.75–81), respectively. To assess the robustness of the models, we performed the evaluation on the unseen test dataset for these chapters and the results showed an accuracy of 0.88, 0.82 and 0.72 for Chapters III, VI and IX, respectively. <b>Interpretation:</b> The present study shows that the combination of panoramic radiograph and clinical oral features could be considered to train a fusion deep learning model for predicting systemic diseases and disorders. |
| first_indexed | 2024-03-09T17:07:59Z |
| format | Article |
| id | doaj.art-01cbaf69b9744b528a327702340274af |
| institution | Directory Open Access Journal |
| issn | 2075-4418 |
| language | English |
| last_indexed | 2024-03-09T17:07:59Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj.art-01cbaf69b9744b528a327702340274af2023-11-24T14:20:20ZengMDPI AGDiagnostics2075-44182022-12-011212319210.3390/diagnostics12123192A Multimodal Deep Learning Approach to Predicting Systemic Diseases from Oral ConditionsDan Zhao0Morteza Homayounfar1Zhe Zhen2Mei-Zhen Wu3Shuk Yin Yu4Kai-Hang Yiu5Varut Vardhanabhuti6George Pelekos7Lijian Jin8Mohamad Koohi-Moghadam9Division of Periodontology & Implant Dentistry, Faculty of Dentistry, The University of Hong Kong, Hong Kong SAR, ChinaDivision of Applied Oral Sciences & Community Dental Care, Faculty of Dentistry, The University of Hong Kong, Hong Kong SAR, ChinaDivision of Cardiology, Department of Medicine, The University of Hong Kong-Shenzhen Hospital, Shenzhen 518009, ChinaDivision of Cardiology, Department of Medicine, The University of Hong Kong-Shenzhen Hospital, Shenzhen 518009, ChinaDivision of Cardiology, Department of Medicine, The University of Hong Kong-Shenzhen Hospital, Shenzhen 518009, ChinaDivision of Cardiology, Department of Medicine, The University of Hong Kong-Shenzhen Hospital, Shenzhen 518009, ChinaDepartment of Diagnostic Radiology, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong SAR, ChinaDivision of Periodontology & Implant Dentistry, Faculty of Dentistry, The University of Hong Kong, Hong Kong SAR, ChinaDivision of Periodontology & Implant Dentistry, Faculty of Dentistry, The University of Hong Kong, Hong Kong SAR, ChinaDivision of Applied Oral Sciences & Community Dental Care, Faculty of Dentistry, The University of Hong Kong, Hong Kong SAR, China<b>Background:</b> It is known that oral diseases such as periodontal (gum) disease are closely linked to various systemic diseases and disorders. Deep learning advances have the potential to make major contributions to healthcare, particularly in the domains that rely on medical imaging. Incorporating non-imaging information based on clinical and laboratory data may allow clinicians to make more comprehensive and accurate decisions. <b>Methods:</b> Here, we developed a multimodal deep learning method to predict systemic diseases and disorders from oral health conditions. A dual-loss autoencoder was used in the first phase to extract periodontal disease-related features from 1188 panoramic radiographs. Then, in the second phase, we fused the image features with the demographic data and clinical information taken from electronic health records (EHR) to predict systemic diseases. We used receiver operation characteristics (ROC) and accuracy to evaluate our model. The model was further validated by an unseen test dataset. <b>Findings:</b> According to our findings, the top three most accurately predicted chapters, in order, are the Chapters III, VI and IX. The results indicated that the proposed model could predict systemic diseases belonging to Chapters III, VI and IX, with AUC values of 0.92 (95% CI, 0.90–94), 0.87 (95% CI, 0.84–89) and 0.78 (95% CI, 0.75–81), respectively. To assess the robustness of the models, we performed the evaluation on the unseen test dataset for these chapters and the results showed an accuracy of 0.88, 0.82 and 0.72 for Chapters III, VI and IX, respectively. <b>Interpretation:</b> The present study shows that the combination of panoramic radiograph and clinical oral features could be considered to train a fusion deep learning model for predicting systemic diseases and disorders.https://www.mdpi.com/2075-4418/12/12/3192periodontal diseasesystemic comorbiditymultimodal deep learningpanoramic radiographelectronic health records |
| spellingShingle | Dan Zhao Morteza Homayounfar Zhe Zhen Mei-Zhen Wu Shuk Yin Yu Kai-Hang Yiu Varut Vardhanabhuti George Pelekos Lijian Jin Mohamad Koohi-Moghadam A Multimodal Deep Learning Approach to Predicting Systemic Diseases from Oral Conditions Diagnostics periodontal disease systemic comorbidity multimodal deep learning panoramic radiograph electronic health records |
| title | A Multimodal Deep Learning Approach to Predicting Systemic Diseases from Oral Conditions |
| title_full | A Multimodal Deep Learning Approach to Predicting Systemic Diseases from Oral Conditions |
| title_fullStr | A Multimodal Deep Learning Approach to Predicting Systemic Diseases from Oral Conditions |
| title_full_unstemmed | A Multimodal Deep Learning Approach to Predicting Systemic Diseases from Oral Conditions |
| title_short | A Multimodal Deep Learning Approach to Predicting Systemic Diseases from Oral Conditions |
| title_sort | multimodal deep learning approach to predicting systemic diseases from oral conditions |
| topic | periodontal disease systemic comorbidity multimodal deep learning panoramic radiograph electronic health records |
| url | https://www.mdpi.com/2075-4418/12/12/3192 |
| work_keys_str_mv | AT danzhao amultimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT mortezahomayounfar amultimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT zhezhen amultimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT meizhenwu amultimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT shukyinyu amultimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT kaihangyiu amultimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT varutvardhanabhuti amultimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT georgepelekos amultimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT lijianjin amultimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT mohamadkoohimoghadam amultimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT danzhao multimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT mortezahomayounfar multimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT zhezhen multimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT meizhenwu multimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT shukyinyu multimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT kaihangyiu multimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT varutvardhanabhuti multimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT georgepelekos multimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT lijianjin multimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions AT mohamadkoohimoghadam multimodaldeeplearningapproachtopredictingsystemicdiseasesfromoralconditions |