Makoto TAKAMATSU – Medium

Makoto TAKAMATSU

Pinned

Published in
Artificial Intelligence in Plain English

Improve Segmentation Accuracy by Correcting for Gaps in Predictions

Silhouette mask segmentation mimics the clinician’s annotation approach

Dec 20, 2020

Improve Segmentation Accuracy by Correcting for Gaps in Predictions

Dec 20, 2020

Pinned

Published in
Towards AI

Application of Uncertainty Modeling to Face Recognition

An facial recognition algorithm that effectively mitigates the negative impact of dirty samples during model training

Dec 3, 2020

Application of Uncertainty Modeling to Face Recognition

Dec 3, 2020

Pinned

[Review] Poor Images Diagnosis [part1]

Towards a reliable medical AI

Nov 19, 2020

[Review] Poor Images Diagnosis [part1]

Nov 19, 2020

Published in
AIメディカルサービス

Bottleneck Transformer

Self attentionをResNetに組み込んだTransformer

Jul 7, 2021

Bottleneck Transformer

Jul 7, 2021

Published in
Geek Culture

Gastric Histopathology Image Classification by Transformer, GasHis-Transformer

CNN and Transformer Fusion for Gastric Histopathology Image Classification with Few False Negatives and False Positive

Jun 21, 2021

Gastric Histopathology Image Classification by Transformer, GasHis-Transformer

Jun 21, 2021

Published in
AIメディカルサービス

Transformerによる胃組織病理画像分類，GasHis-Transformer

CNNとTransformerの融合で偽陰性の少ない胃組織病理学的画像分類を実現

Jun 20, 2021

Transformerによる胃組織病理画像分類，GasHis-Transformer

Jun 20, 2021

Published in
AIメディカルサービス

[Transformer &物体検知] You Only Look at One Sequence (YOLOS)

画像識別タスクで事前学習したTransformerを物体検出タスクに転移学習ができる！？

Jun 9, 2021

[Transformer &物体検知] You Only Look at One Sequence (YOLOS)

Jun 9, 2021

Published in
AIメディカルサービス

[論文紹介] GANを使用しない効果的なドメイン適応

教師なしドメイン適応，ソース分布とターゲット分布間の不一致を低周波スペクトル交換で減少

May 26, 2021

[論文紹介] GANを使用しない効果的なドメイン適応

May 26, 2021

Published in
Towards AI

A Useful New Image Classification Method That Uses neither CNNs nor Attention

[Paper summury] MLP-mix

May 19, 2021

A Useful New Image Classification Method That Uses neither CNNs nor Attention

May 19, 2021

Published in
AIメディカルサービス

[論文紹介] MLP-Mixer: An all-MLP Architecture for Vision

CNNもAttentionも使わない有用な画像分類手法

May 19, 2021

[論文紹介] MLP-Mixer: An all-MLP Architecture for Vision

May 19, 2021

Makoto TAKAMATSU

Makoto TAKAMATSU

Medical AI Engineer, My main interest is to improve the efficiency of medical diagnosis. my LinkedIn: https://www.linkedin.com/in/makoto-takamatsu-8b1884141/

Following

Help
Status
About
Careers
Press
Blog
Privacy
Terms
Text to speech
Teams