Deep-learning-based, computer-aided classifier developed with a small dataset of clinical images surpasses board-certified dermatologists in skin tumour diagnosis

BACKGROUND

Application of deep-learning technology to skin cancer classification can potentially improve the sensitivity and specificity of skin cancer screening, but the number of training images required for such a system is thought to be extremely large.

OBJECTIVES

To determine whether deep-learning technology could be used to develop an efficient skin cancer classification system with a relatively small dataset of clinical images.

METHODS

A deep convolutional neural network (DCNN) was trained using a dataset of 4867 clinical images obtained from 1842 patients diagnosed with skin tumours at the University of Tsukuba Hospital from 2003 to 2016. The images consisted of 14 diagnoses, including both malignant and benign conditions. Its performance was tested against 13 board-certified dermatologists and nine dermatology trainees.

RESULTS

The overall classification accuracy of the trained DCNN was 76·5%. The DCNN achieved 96·3% sensitivity (correctly classified malignant as malignant) and 89·5% specificity (correctly classified benign as benign). Although the accuracy of malignant or benign classification by the board-certified dermatologists was statistically higher than that of the dermatology trainees (85·3% ± 3·7% and 74·4% ± 6·8%, P < 0·01), the DCNN achieved even greater accuracy, as high as 92·4% ± 2·1% (P < 0·001).

CONCLUSIONS

We have developed an efficient skin tumour classifier using a DCNN trained on a relatively small dataset. The DCNN classified images of skin tumours more accurately than board-certified dermatologists. Collectively, the current system may have capabilities for screening purposes in general medical practice, particularly because it requires only a single clinical image for classification.

Intravenous immunoglobulin contributes to the control of antimelanoma differentiation-associated protein 5 antibody-associated dermatomyositis with palmar violaceous macules/papules

Autoantibodies to melanoma differentiation-associated protein 5 (MDA5) are associated with a subset of patients with dermatomyositis (DM) who have rapidly progressive interstitial lung disease (RP-ILD) with poor prognosis. Intensive immunosuppressive therapy is initiated before irreversible lung damage can occur; however, there are few lines of evidence for the treatment of RP-ILD. Here, we report three cases of anti-MDA5 antibody-associated DM with RP-ILD in which the patients were treated with combined-modality therapy, including high-dose prednisolone, tacrolimus, intravenous cyclophosphamide and intravenous immunoglobulin (IVIG). In all three cases, serum ferritin levels, which are known to represent the disease activity of RP-ILD, were decreased after IVIG administration. IVIG might contribute to the control of the disease activity of anti-MDA5 antibody-positive DM. Moreover, palmar violaceous macules/papules around the interphalangeal joints, which was observed in all three cases in the incipient stage, might be a useful sign in suggesting a diagnosis of anti-MDA5 antibody-associated DM.

Fluorogenic and fluorescent labeling reagents with a benzofurazan skeleton

Fluorogenic and fluorescent labeling reagents having a benzofurazan (2,1,3-benzoxadiazole) skeleton such as 4-fluoro-7-nitro-2,1,3-benzoxadiazole (NBD-F), 4-N,N-dimethylaminosulfonyl-7-fluoro-2,1,3-benzoxadiazole (DBD-F), 4-aminosulfonyl-7-fluoro-2,1,3-benzoxadiazole (ABD-F), ammonium 7-fluoro-2,1,3-benzoxadiazole-4-sulfonate (SBD-F), 4-hydrazino-7-nitro-2,1,3-benzoxadiazole (NBD-H), 4-N,N-dimethylaminosulfonyl-7-hydrazino-2,1,3-benzoxadiazole (DBD-H), 4-nitro-7-N-piperazino-2,1,3-benzoxadiazole (NBD-PZ), 4-N,N-dimethylaminosulfonyl-7-N-piperazino-2,1,3-benzoxadiazole (DBD-PZ), 4-(N-chloroformylmethyl-N-methyl)amino-7-N,N-dimethylaminosulfonyl-2,1,3-benzoxadiazole (DBD-COCl) and 7-N,N-dimethylaminosulfonyl-4-(2,1,3-benzoxadiazolyl) isothiocyanate (DBD-NCS) are reviewed in terms of synthetic method, reactivity, fluorescence characteristics, sensitivity and application to analytes.