Deep Learning for Diagnostic Binary Classification of Multiple-Lesion Skin Diseases
Background: Diagnosis of skin diseases is often challenging and computer-aided diagnostic tools are urgently needed to underpin decision making.Objective: To develop a convolutional neural network model to classify clinically relevant selected multiple-lesion skin diseases, this in accordance to the...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2020-09-01
|
| Series: | Frontiers in Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/article/10.3389/fmed.2020.574329/full |
| _version_ | 1828866274442084352 |
|---|---|
| author | Kenneth Thomsen Anja Liljedahl Christensen Lars Iversen Hans Bredsted Lomholt Ole Winther Ole Winther Ole Winther |
| author_facet | Kenneth Thomsen Anja Liljedahl Christensen Lars Iversen Hans Bredsted Lomholt Ole Winther Ole Winther Ole Winther |
| author_sort | Kenneth Thomsen |
| collection | DOAJ |
| description | Background: Diagnosis of skin diseases is often challenging and computer-aided diagnostic tools are urgently needed to underpin decision making.Objective: To develop a convolutional neural network model to classify clinically relevant selected multiple-lesion skin diseases, this in accordance to the STARD guidelines.Methods: This was an image-based retrospective study using multi-task learning for binary classification. A VGG-16 model was trained on 16,543 non-standardized images. Image data was distributed in training set (80%), validation set (10%), and test set (10%). All images were collected from a clinical database of a Danish population attending one dermatological department. Included was patients categorized with ICD-10 codes related to acne, rosacea, psoriasis, eczema, and cutaneous t-cell lymphoma.Results: Acne was distinguished from rosacea with a sensitivity of 85.42% CI 72.24–93.93% and a specificity of 89.53% CI 83.97–93.68%, cutaneous t-cell lymphoma was distinguished from eczema with a sensitivity of 74.29% CI 67.82–80.05% and a specificity of 84.09% CI 80.83–86.99%, and psoriasis from eczema with a sensitivity of 81.79% CI 78.51–84.76% and a specificity of 73.57% CI 69.76–77.13%. All results were based on the test set.Conclusion: The performance rates reported were equal or superior to those reported for general practitioners with dermatological training, indicating that computer-aided diagnostic models based on convolutional neural network may potentially be employed for diagnosing multiple-lesion skin diseases. |
| first_indexed | 2024-12-13T04:46:29Z |
| format | Article |
| id | doaj.art-e16b45824a994825bbba29ca10318e6c |
| institution | Directory Open Access Journal |
| issn | 2296-858X |
| language | English |
| last_indexed | 2024-12-13T04:46:29Z |
| publishDate | 2020-09-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Medicine |
| spelling | doaj.art-e16b45824a994825bbba29ca10318e6c2022-12-21T23:59:08ZengFrontiers Media S.A.Frontiers in Medicine2296-858X2020-09-01710.3389/fmed.2020.574329574329Deep Learning for Diagnostic Binary Classification of Multiple-Lesion Skin DiseasesKenneth Thomsen0Anja Liljedahl Christensen1Lars Iversen2Hans Bredsted Lomholt3Ole Winther4Ole Winther5Ole Winther6Department of Dermatology and Venereology, Aarhus University Hospital, Aarhus, DenmarkDepartment of Applied Mathematics and Computer Science, Technical University of Denmark, Lyngby, DenmarkDepartment of Dermatology and Venereology, Aarhus University Hospital, Aarhus, DenmarkClinical Institute, Aalborg University, Aalborg, DenmarkDepartment of Applied Mathematics and Computer Science, Technical University of Denmark, Lyngby, DenmarkCenter for Genomic Medicine, Rigshospitalet, Copenhagen University Hospital, Copenhagen, DenmarkDepartment of Biology, Bioinformatics Centre, University of Copenhagen, Copenhagen, DenmarkBackground: Diagnosis of skin diseases is often challenging and computer-aided diagnostic tools are urgently needed to underpin decision making.Objective: To develop a convolutional neural network model to classify clinically relevant selected multiple-lesion skin diseases, this in accordance to the STARD guidelines.Methods: This was an image-based retrospective study using multi-task learning for binary classification. A VGG-16 model was trained on 16,543 non-standardized images. Image data was distributed in training set (80%), validation set (10%), and test set (10%). All images were collected from a clinical database of a Danish population attending one dermatological department. Included was patients categorized with ICD-10 codes related to acne, rosacea, psoriasis, eczema, and cutaneous t-cell lymphoma.Results: Acne was distinguished from rosacea with a sensitivity of 85.42% CI 72.24–93.93% and a specificity of 89.53% CI 83.97–93.68%, cutaneous t-cell lymphoma was distinguished from eczema with a sensitivity of 74.29% CI 67.82–80.05% and a specificity of 84.09% CI 80.83–86.99%, and psoriasis from eczema with a sensitivity of 81.79% CI 78.51–84.76% and a specificity of 73.57% CI 69.76–77.13%. All results were based on the test set.Conclusion: The performance rates reported were equal or superior to those reported for general practitioners with dermatological training, indicating that computer-aided diagnostic models based on convolutional neural network may potentially be employed for diagnosing multiple-lesion skin diseases.https://www.frontiersin.org/article/10.3389/fmed.2020.574329/fulldeep neural network (DNN)dermatologyskin diseaseacnerosaceapsoriasis |
| spellingShingle | Kenneth Thomsen Anja Liljedahl Christensen Lars Iversen Hans Bredsted Lomholt Ole Winther Ole Winther Ole Winther Deep Learning for Diagnostic Binary Classification of Multiple-Lesion Skin Diseases Frontiers in Medicine deep neural network (DNN) dermatology skin disease acne rosacea psoriasis |
| title | Deep Learning for Diagnostic Binary Classification of Multiple-Lesion Skin Diseases |
| title_full | Deep Learning for Diagnostic Binary Classification of Multiple-Lesion Skin Diseases |
| title_fullStr | Deep Learning for Diagnostic Binary Classification of Multiple-Lesion Skin Diseases |
| title_full_unstemmed | Deep Learning for Diagnostic Binary Classification of Multiple-Lesion Skin Diseases |
| title_short | Deep Learning for Diagnostic Binary Classification of Multiple-Lesion Skin Diseases |
| title_sort | deep learning for diagnostic binary classification of multiple lesion skin diseases |
| topic | deep neural network (DNN) dermatology skin disease acne rosacea psoriasis |
| url | https://www.frontiersin.org/article/10.3389/fmed.2020.574329/full |
| work_keys_str_mv | AT kenneththomsen deeplearningfordiagnosticbinaryclassificationofmultiplelesionskindiseases AT anjaliljedahlchristensen deeplearningfordiagnosticbinaryclassificationofmultiplelesionskindiseases AT larsiversen deeplearningfordiagnosticbinaryclassificationofmultiplelesionskindiseases AT hansbredstedlomholt deeplearningfordiagnosticbinaryclassificationofmultiplelesionskindiseases AT olewinther deeplearningfordiagnosticbinaryclassificationofmultiplelesionskindiseases AT olewinther deeplearningfordiagnosticbinaryclassificationofmultiplelesionskindiseases AT olewinther deeplearningfordiagnosticbinaryclassificationofmultiplelesionskindiseases |