Submitted by kej2006 on June 6, 2018 - 4:11pm
| Title | Digital imaging biomarkers feed machine learning for melanoma screening. |
| Publication Type | Journal Article |
| Year of Publication | 2017 |
| Authors | Gareau DS, Da Rosa JCorrea, Yagerman S, Carucci JA, Gulati N, Hueto F, DeFazio JL, Suárez-Fariñas M, Marghoob A, Krueger JG |
| Journal | Exp Dermatol |
| Volume | 26 |
| Issue | 7 |
| Pagination | 615-618 |
| Date Published | 2017 07 |
| ISSN | 1600-0625 |
| Keywords | Algorithms, Automation, Biomarkers, Tumor, Color, Dermatology, Dermoscopy, Diagnosis, Differential, Dysplastic Nevus Syndrome, Humans, Image Processing, Computer-Assisted, Machine Learning, Melanoma, Nevus, Pigmented, Pattern Recognition, Automated, Pigmentation, Reproducibility of Results, Risk, Sensitivity and Specificity, Skin Neoplasms |
| Abstract | We developed an automated approach for generating quantitative image analysis metrics (imaging biomarkers) that are then analysed with a set of 13 machine learning algorithms to generate an overall risk score that is called a Q-score. These methods were applied to a set of 120 "difficult" dermoscopy images of dysplastic nevi and melanomas that were subsequently excised/classified. This approach yielded 98% sensitivity and 36% specificity for melanoma detection, approaching sensitivity/specificity of expert lesion evaluation. Importantly, we found strong spectral dependence of many imaging biomarkers in blue or red colour channels, suggesting the need to optimize spectral evaluation of pigmented lesions. |
| DOI | 10.1111/exd.13250 |
| Alternate Journal | Exp. Dermatol. |
| PubMed ID | 27783441 |
| PubMed Central ID | PMC5516237 |
| Grant List | R01 CA193390 / CA / NCI NIH HHS / United States T32 GM007739 / GM / NIGMS NIH HHS / United States UL1 TR001866 / TR / NCATS NIH HHS / United States |