Systematic review of deep learning image analyses for the diagnosis and monitoring of skin disease
Abstract Skin diseases affect one-third of the global population, posing a major healthcare burden. Deep learning may optimise healthcare workflows through processing skin images via neural networks to make predictions. A focus of deep learning research is skin lesion triage to detect cancer, but th...
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-09-01
|
| Series: | npj Digital Medicine |
| Online Access: | https://doi.org/10.1038/s41746-023-00914-8 |
| _version_ | 1797556238913896448 |
|---|---|
| author | Shern Ping Choy Byung Jin Kim Alexandra Paolino Wei Ren Tan Sarah Man Lin Lim Jessica Seo Sze Ping Tan Luc Francis Teresa Tsakok Michael Simpson Jonathan N. W. N. Barker Magnus D. Lynch Mark S. Corbett Catherine H. Smith Satveer K. Mahil |
| author_facet | Shern Ping Choy Byung Jin Kim Alexandra Paolino Wei Ren Tan Sarah Man Lin Lim Jessica Seo Sze Ping Tan Luc Francis Teresa Tsakok Michael Simpson Jonathan N. W. N. Barker Magnus D. Lynch Mark S. Corbett Catherine H. Smith Satveer K. Mahil |
| author_sort | Shern Ping Choy |
| collection | DOAJ |
| description | Abstract Skin diseases affect one-third of the global population, posing a major healthcare burden. Deep learning may optimise healthcare workflows through processing skin images via neural networks to make predictions. A focus of deep learning research is skin lesion triage to detect cancer, but this may not translate to the wider scope of >2000 other skin diseases. We searched for studies applying deep learning to skin images, excluding benign/malignant lesions (1/1/2000-23/6/2022, PROSPERO CRD42022309935). The primary outcome was accuracy of deep learning algorithms in disease diagnosis or severity assessment. We modified QUADAS-2 for quality assessment. Of 13,857 references identified, 64 were included. The most studied diseases were acne, psoriasis, eczema, rosacea, vitiligo, urticaria. Deep learning algorithms had high specificity and variable sensitivity in diagnosing these conditions. Accuracy of algorithms in diagnosing acne (median 94%, IQR 86–98; n = 11), rosacea (94%, 90–97; n = 4), eczema (93%, 90–99; n = 9) and psoriasis (89%, 78–92; n = 8) was high. Accuracy for grading severity was highest for psoriasis (range 93–100%, n = 2), eczema (88%, n = 1), and acne (67–86%, n = 4). However, 59 (92%) studies had high risk-of-bias judgements and 62 (97%) had high-level applicability concerns. Only 12 (19%) reported participant ethnicity/skin type. Twenty-four (37.5%) evaluated the algorithm in an independent dataset, clinical setting or prospectively. These data indicate potential of deep learning image analysis in diagnosing and monitoring common skin diseases. Current research has important methodological/reporting limitations. Real-world, prospectively-acquired image datasets with external validation/testing will advance deep learning beyond the current experimental phase towards clinically-useful tools to mitigate rising health and cost impacts of skin disease. |
| first_indexed | 2024-03-10T16:59:57Z |
| format | Article |
| id | doaj.art-7895d670344740b89be10ad97311a1b5 |
| institution | Directory Open Access Journal |
| issn | 2398-6352 |
| language | English |
| last_indexed | 2024-03-10T16:59:57Z |
| publishDate | 2023-09-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Digital Medicine |
| spelling | doaj.art-7895d670344740b89be10ad97311a1b52023-11-20T11:01:05ZengNature Portfolionpj Digital Medicine2398-63522023-09-016111110.1038/s41746-023-00914-8Systematic review of deep learning image analyses for the diagnosis and monitoring of skin diseaseShern Ping Choy0Byung Jin Kim1Alexandra Paolino2Wei Ren Tan3Sarah Man Lin Lim4Jessica Seo5Sze Ping Tan6Luc Francis7Teresa Tsakok8Michael Simpson9Jonathan N. W. N. Barker10Magnus D. Lynch11Mark S. Corbett12Catherine H. Smith13Satveer K. Mahil14St John’s Institute of Dermatology, Guy’s and St Thomas’ NHS Foundation Trust and King’s College LondonSt George’s University Hospitals NHS Foundation TrustSt John’s Institute of Dermatology, Guy’s and St Thomas’ NHS Foundation Trust and King’s College LondonSt John’s Institute of Dermatology, Guy’s and St Thomas’ NHS Foundation Trust and King’s College LondonMaidstone and Tunbridge Wells NHS TrustImperial College LondonBarking, Havering and Redbridge University Hospitals NHS TrustSt John’s Institute of Dermatology, Guy’s and St Thomas’ NHS Foundation Trust and King’s College LondonSt John’s Institute of Dermatology, Guy’s and St Thomas’ NHS Foundation Trust and King’s College LondonDepartment of Medical and Molecular Genetics, King’s College LondonSt John’s Institute of Dermatology, Guy’s and St Thomas’ NHS Foundation Trust and King’s College LondonSt John’s Institute of Dermatology, Guy’s and St Thomas’ NHS Foundation Trust and King’s College LondonCenter for Reviews and Dissemination, University of YorkSt John’s Institute of Dermatology, Guy’s and St Thomas’ NHS Foundation Trust and King’s College LondonSt John’s Institute of Dermatology, Guy’s and St Thomas’ NHS Foundation Trust and King’s College LondonAbstract Skin diseases affect one-third of the global population, posing a major healthcare burden. Deep learning may optimise healthcare workflows through processing skin images via neural networks to make predictions. A focus of deep learning research is skin lesion triage to detect cancer, but this may not translate to the wider scope of >2000 other skin diseases. We searched for studies applying deep learning to skin images, excluding benign/malignant lesions (1/1/2000-23/6/2022, PROSPERO CRD42022309935). The primary outcome was accuracy of deep learning algorithms in disease diagnosis or severity assessment. We modified QUADAS-2 for quality assessment. Of 13,857 references identified, 64 were included. The most studied diseases were acne, psoriasis, eczema, rosacea, vitiligo, urticaria. Deep learning algorithms had high specificity and variable sensitivity in diagnosing these conditions. Accuracy of algorithms in diagnosing acne (median 94%, IQR 86–98; n = 11), rosacea (94%, 90–97; n = 4), eczema (93%, 90–99; n = 9) and psoriasis (89%, 78–92; n = 8) was high. Accuracy for grading severity was highest for psoriasis (range 93–100%, n = 2), eczema (88%, n = 1), and acne (67–86%, n = 4). However, 59 (92%) studies had high risk-of-bias judgements and 62 (97%) had high-level applicability concerns. Only 12 (19%) reported participant ethnicity/skin type. Twenty-four (37.5%) evaluated the algorithm in an independent dataset, clinical setting or prospectively. These data indicate potential of deep learning image analysis in diagnosing and monitoring common skin diseases. Current research has important methodological/reporting limitations. Real-world, prospectively-acquired image datasets with external validation/testing will advance deep learning beyond the current experimental phase towards clinically-useful tools to mitigate rising health and cost impacts of skin disease.https://doi.org/10.1038/s41746-023-00914-8 |
| spellingShingle | Shern Ping Choy Byung Jin Kim Alexandra Paolino Wei Ren Tan Sarah Man Lin Lim Jessica Seo Sze Ping Tan Luc Francis Teresa Tsakok Michael Simpson Jonathan N. W. N. Barker Magnus D. Lynch Mark S. Corbett Catherine H. Smith Satveer K. Mahil Systematic review of deep learning image analyses for the diagnosis and monitoring of skin disease npj Digital Medicine |
| title | Systematic review of deep learning image analyses for the diagnosis and monitoring of skin disease |
| title_full | Systematic review of deep learning image analyses for the diagnosis and monitoring of skin disease |
| title_fullStr | Systematic review of deep learning image analyses for the diagnosis and monitoring of skin disease |
| title_full_unstemmed | Systematic review of deep learning image analyses for the diagnosis and monitoring of skin disease |
| title_short | Systematic review of deep learning image analyses for the diagnosis and monitoring of skin disease |
| title_sort | systematic review of deep learning image analyses for the diagnosis and monitoring of skin disease |
| url | https://doi.org/10.1038/s41746-023-00914-8 |
| work_keys_str_mv | AT shernpingchoy systematicreviewofdeeplearningimageanalysesforthediagnosisandmonitoringofskindisease AT byungjinkim systematicreviewofdeeplearningimageanalysesforthediagnosisandmonitoringofskindisease AT alexandrapaolino systematicreviewofdeeplearningimageanalysesforthediagnosisandmonitoringofskindisease AT weirentan systematicreviewofdeeplearningimageanalysesforthediagnosisandmonitoringofskindisease AT sarahmanlinlim systematicreviewofdeeplearningimageanalysesforthediagnosisandmonitoringofskindisease AT jessicaseo systematicreviewofdeeplearningimageanalysesforthediagnosisandmonitoringofskindisease AT szepingtan systematicreviewofdeeplearningimageanalysesforthediagnosisandmonitoringofskindisease AT lucfrancis systematicreviewofdeeplearningimageanalysesforthediagnosisandmonitoringofskindisease AT teresatsakok systematicreviewofdeeplearningimageanalysesforthediagnosisandmonitoringofskindisease AT michaelsimpson systematicreviewofdeeplearningimageanalysesforthediagnosisandmonitoringofskindisease AT jonathannwnbarker systematicreviewofdeeplearningimageanalysesforthediagnosisandmonitoringofskindisease AT magnusdlynch systematicreviewofdeeplearningimageanalysesforthediagnosisandmonitoringofskindisease AT markscorbett systematicreviewofdeeplearningimageanalysesforthediagnosisandmonitoringofskindisease AT catherinehsmith systematicreviewofdeeplearningimageanalysesforthediagnosisandmonitoringofskindisease AT satveerkmahil systematicreviewofdeeplearningimageanalysesforthediagnosisandmonitoringofskindisease |