An explainable deep learning model classifies eight categories of pigmented skin lesions on clinical photographs: A multicenter retrospective internal validation study in Japan

Kimi Iinuma, Kazuyasu Fujii, Chisa Nakashima, Kenichiro Kasai, Hiroyuki Irie, Hitonari Kanetomo, Shigeto Yanagihara, Sayuri Sato, Hisashi Uhara, Fumiaki Takeda, Yuichi Kimura, Takashi Nagaoka, Atsushi Otsuka · 2026

Read An explainable deep learning model classifies eight categories of pigmented skin lesions on clinical photographs: A multicenter retrospective internal validation study in Japan by Kimi Iinuma, Kazuyasu Fujii, Chisa Nakashima, Kenichiro Kasai, Hiroyuki Irie, Hitonari Kanetomo, Shigeto Yanagihara, Sayuri Sato, Hisashi Uhara, Fumiaki Takeda, Yuichi Kimura, Takashi Nagaoka, Atsushi Otsuka (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13499408.txt.

Description

Kimi Iinuma and colleagues present in 2026 a deep learning system that sorts pigmented skin spots into eight categories from ordinary clinical photographs taken in Japan. Sorting such spots by eye is hard, especially where a magnified skin exam tool is unavailable. The system was trained and checked on 979 photographs gathered from dermatology centres, covering harmless marks such as moles and freckles alongside two skin cancers.

Data were split by patient so no person's images leaked between training and test sets, keeping the check honest. On the held-out test set the model classified most images correctly, reaching accuracy in the mid-ninety percent range in the reported figures. Every melanoma in that small test set was caught, though the authors caution the set was too small to prove cancer safety, and mistakes clustered between look-alike harmless and cancerous growths.

Catalog record

CREATORKimi Iinuma, Kazuyasu Fujii, Chisa Nakashima, Kenichiro Kasai, Hiroyuki Irie, Hitonari Kanetomo, Shigeto Yanagihara, Sayuri Sato, Hisashi Uhara, Fumiaki Takeda, Yuichi Kimura, Takashi Nagaoka, Atsushi Otsuka
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE6065
LICENSEhttps://creativecommons.org/licenses/by/4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/article/MED/PMC13499408

Browse the full library · About the Tower of Babel