Table 1.
Diagnostic accuracy of teledermatology for skin cancer diagnosis
| Source | Population | Design | Outcome |
|---|---|---|---|
| Warshaw et al., [38] | 2152 patients: 2082 male, 70 female (mean age: 68 y) |
Teledermatologist diagnosis using camera or dermoscopy taken images compared to in-person FTF* dermatologist diagnosis (Level: Benign vs. malignant) |
Primary diagnosis agreement between TD** and FTF diagnosis: 45.7–75.7% |
| Borve et al., 2015, [63] | 772 patients: 474 female, 298 male (mean age: 54 y) |
Diagnosis as malignant or benign by 3 teledermatologists using dermoscopy images and clinical information on iDoc24 app compared with histopathological diagnosis (Level: Benign vs. malignant) |
TD benign diagnosis agreed with Histopathology benign diagnosis: 99.1% TD malignant diagnosis agreed with Histopathology malignant diagnosis: 74% |
| Congalton et al., 2015, [64] | 310 patients: 168 female, 142 male (mean age: 58 y) |
Melanoma or non-melanoma diagnosis using MoleMapDiagnose software compared with histological diagnosis (Level: Malignant melanocytic vs. benign) |
PPV*** of teledermatoscopic diagnosis of melanoma and positive histology was 63% |
| Borve et al., 2013, [65] | 62 patients: 24 female, 38 male (mean age: 64 y) |
Interobserver concordance between 2 teledermoscopists and a dermatologist using dermoscopy images and clinical information on iDoc24 smartphone app and FTF diagnosis compared with histopathological diagnosis (Level: Primary diagnosis) |
Teledermoscopic diagnostic accuracy was 50.7% (teledermatologist 1) and 60.9% (teledermatologist 2) as compared with histopathological diagnosis FTF diagnostic accuracy was 66.7% as compared with histopathological diagnosis |
| Borve et al., 2013, [65] | 40 patients: 23 female, 17 male (mean age: 49 y) | Diagnosis by dermatologists using images taken by GP**** from a mobile phone camera compared with FTF diagnosis by the same dermatologists (Level: Benign vs. malignant) | TD diagnosis agreed with FTF diagnosis: 31/40 patients; 78% |
| Boyce et al., 2011, [43] | 55 patients: 22 female, 33 male (mean age: 26 y) |
Dermatologist remote diagnosis using patient-generated images taken by mobile phones compared with traditional FTF diagnosis (Level: Primary management outcome – immediate action required, follow-up in 3 months, or no further action) |
Exact diagnostic agreement in 116/167 analyzed lesions (69%) between remote diagnosis of patient-generated clinical images and FTF diagnosis |
| Warshaw et al., [41] | 542 patients: 23 female, 519 male (mean age: 66 y) |
Store-and-forward teledermatology diagnosis and FTF dermatologist diagnosis compared with histopathology diagnosis (Level: Primary diagnosis) |
TD diagnostic agreement with histopathology diagnosis: 67% FTF diagnostic agreement with histopathology diagnosis: 81% |
*FTF signifies face to Face
**TD signifies Teledermatology
***PPV signifies positive predictive value
**** GP signifies General Practitioner