Skip to content

Our manuscript, "Interpretation of the Outputs of Deep Learning Model Trained with Skin Cancer Dataset" was published as a letter article in the Journal of Investigative Dermatology today (https://www.jidonline.org/article/S0022-202X(18)31992-4/fulltext).

When we train a CNN model, we somtimes get a disappointing Top-1 accuracy. I also suffered this problem and I did not understand exactly what was wrong at that time. When my early version of the 12DX paper was reviewed in JAMA dermatology 2 years ago, the biggest reason for rejection was the low Top-1 accuracy.


However, unlike general object recognition studies, it is very difficult to determine medical research results with Top-1 accuracy, and it is important that the AUC can be high even with a low Top-1 accuracy. If you look carefully, most of medical AI researches have used AUC rather than Top-(n) accuracy.

Because of small and imbalanced training data in medical researches, the analysis of each class as Top-(n) accuracy is inadequate (but the mean Top-(n) of all classes is meaningful). Top-(n) accuracy of each classes vary whenever we repeat the training of CNN with imbalanced dataset. Therefore, we should see the corrected value while using thresholds of each classes, that is ROC curve.

With the AUC results, we published "Classification of the Clinical Images for Benign and Malignant Cutaneous Tumors Using a Deep Learning Algorithm" (https://www.jidonline.org/article/S0022-202X(18)30111-8/fulltext)

There was a debate that my 12DX algorithm is not sensitive (low top-1 accuracy) with the ISIC dataset (Automated Dermatological Diagnosis: Hype or Reality?; https://www.jidonline.org/article/S0022-202X(18)31991-2/fulltext).

 

There was an additional problem as well as the Top accuracy problem.

When we analyze a clinical image, "the problem of judging whether it is melanoma or not" is easier than "the problem of matching the type of cancer".

Analyzing the output of the AI ​​(CNN) model is equivalent to "the problem of matching the type of cancer", and analyzing the ratio of output is proper if we want to analyze the problem of judging "whether cancer or not".

We interpreted the ratio of melanoma output and nevus output rather than using melanoma output alone.

RATIO (Melanoma Index) = melanoma output / (melanoma output + nevus output).

The clinical image of skin cancer consists of a nodular lesion and a background. If you want to concentrate on only the lesion, we need to analyze it with RATIO as above to get more accurate results.

In the attached photograph, (b) is "matching what cancer is" and (a) is judging "whether it is cancer or not".

We made web-DEMO (http://dx.medicalphoto.org), and we have made it possible to show what conclusions are coming up depending on the Top-5 output and how it is interpreted.

 



번호 제목 글쓴이 날짜 조회 수
289 reddit downloader WHRIA 2020.06.21 220
288 steamlit WHRIA 2020.10.15 220
287 전세금 돌려받는 법 WHRIA 2019.02.11 219
286 fda 인증 WHRIA 2020.09.03 219
285 리눅스 samba mount 정리 [1] WHRIA 2019.05.26 200
284 excel 에서 소수점 표시 함수 WHRIA 2018.11.08 197
283 airplane WHRIA 2020.02.10 197
282 mysql python reconnect [2] WHRIA 2020.04.24 196
281 sql python WHRIA 2020.07.01 195
280 고등학교 친구 하나가 죽었다던데... WHRIA 2015.01.18 188
279 caffe hdf5 error WHRIA 2020.06.22 187
278 진료실 녹음 스크립트 file WHRIA 2015.12.02 180
277 regression model [6] WHRIA 2020.06.18 175
276 faster rcnn resnet 101 WHRIA 2018.09.09 169
275 usb 3.1 + DP WHRIA 2020.01.07 168

Powered by Xpress Engine / Designed by Sketchbook

sketchbook5, 스케치북5

sketchbook5, 스케치북5

나눔글꼴 설치 안내


이 PC에는 나눔글꼴이 설치되어 있지 않습니다.

이 사이트를 나눔글꼴로 보기 위해서는
나눔글꼴을 설치해야 합니다.

설치 취소