2012 Fiscal Year Final Research Report
Development of an automated melanoma discrimination system
Project/Area Number |
23791295
|
Research Category |
Grant-in-Aid for Young Scientists (B)
|
Allocation Type | Multi-year Fund |
Research Field |
Dermatology
|
Research Institution | Hosei University |
Principal Investigator |
|
Project Period (FY) |
2011 – 2012
|
Keywords | 自動診断支援 / ダーモスコピー / メラノーマ / 画像解析 |
Research Abstract |
We developed a new automated melanoma discrimination system accommodating a wide variety of pigmented skin lesions including not only melanocytic skin lesions (MSLs), but also non-melanocytic skin lesions (noMSLs). Our new system was developed on a total 658 dermoscopy images (42 melanomas, 506 benign MSLs and 110 noMSLs) and achieved a sensitivity of 97.6% and a specificity of 92.2% under the leave-one-out cross-validation test.
|
-
-
-
-
-
-
-
-
-
[Presentation] 3D Architecture of Ductal Carcinoma in Situ From Image Reconstructions2012
Author(s)
Kerri-Ann Norton, Sameera Namazi, Nicola Barnard, Mariko Fujibayashi, Gyan Bhanot, Shridar Ganesan, Hitoshi Iyatomi, Koichi Ogawa, and Troy Shinbrot
Organizer
Proc. IEEE EMBS Conference on Biomedical Engineering (IECBES 2012)
Place of Presentation
Langkawi, Malaysia,
Year and Date
20121217-1219
-
-
-
-
-
-