Few-Shot Learning for Dermatological Disease Diagnosis

Viraj Prabhu, Anitha Kannan, Murali Ravuri, Manish Chaplain, David Sontag, Xavier Amatriain
; Proceedings of the 4th Machine Learning for Healthcare Conference, PMLR 106:532-552, 2019.

Abstract

We consider the problem of clinical image classification for the purpose of aiding doctors in dermatological disease diagnosis. Diagnosis of dermatological conditions from images poses two major challenges for standard off-the-shelf techniques: First, the distribution of real-world dermatological datasets is typically long-tailed. Second, intra-class variability is large. To address the first issue, we formulate the problem as low-shot learning, where once deployed, a base classifier must rapidly generalize to diagnose novel conditions given very few labeled examples. To model intra-class variability effectively, we propose Prototypical Clustering Networks (PCN), an extension to Prototypical Networks (Snell et al. , 2017 ) that learns a mixture of “prototypes” for each class. Prototypes are initialized for each class via clustering and refined via an online update scheme. Classification is performed by measuring similarity to a weighted combination of prototypes within a class, where the weights are the inferred cluster responsibilities. We demonstrate the strengths of our approach in effective diagnosis on a realistic dataset of dermatological conditions.

Cite this Paper


BibTeX
@InProceedings{pmlr-v106-prabhu19a, title = {Few-Shot Learning for Dermatological Disease Diagnosis}, author = {Prabhu, Viraj and Kannan, Anitha and Ravuri, Murali and Chaplain, Manish and Sontag, David and Amatriain, Xavier}, booktitle = {Proceedings of the 4th Machine Learning for Healthcare Conference}, pages = {532--552}, year = {2019}, editor = {Finale Doshi-Velez and Jim Fackler and Ken Jung and David Kale and Rajesh Ranganath and Byron Wallace and Jenna Wiens}, volume = {106}, series = {Proceedings of Machine Learning Research}, address = {Ann Arbor, Michigan}, month = {09--10 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v106/prabhu19a/prabhu19a.pdf}, url = {http://proceedings.mlr.press/v106/prabhu19a.html}, abstract = {We consider the problem of clinical image classification for the purpose of aiding doctors in dermatological disease diagnosis. Diagnosis of dermatological conditions from images poses two major challenges for standard off-the-shelf techniques: First, the distribution of real-world dermatological datasets is typically long-tailed. Second, intra-class variability is large. To address the first issue, we formulate the problem as low-shot learning, where once deployed, a base classifier must rapidly generalize to diagnose novel conditions given very few labeled examples. To model intra-class variability effectively, we propose Prototypical Clustering Networks (PCN), an extension to Prototypical Networks (Snell et al. , 2017 ) that learns a mixture of “prototypes” for each class. Prototypes are initialized for each class via clustering and refined via an online update scheme. Classification is performed by measuring similarity to a weighted combination of prototypes within a class, where the weights are the inferred cluster responsibilities. We demonstrate the strengths of our approach in effective diagnosis on a realistic dataset of dermatological conditions.} }
Endnote
%0 Conference Paper %T Few-Shot Learning for Dermatological Disease Diagnosis %A Viraj Prabhu %A Anitha Kannan %A Murali Ravuri %A Manish Chaplain %A David Sontag %A Xavier Amatriain %B Proceedings of the 4th Machine Learning for Healthcare Conference %C Proceedings of Machine Learning Research %D 2019 %E Finale Doshi-Velez %E Jim Fackler %E Ken Jung %E David Kale %E Rajesh Ranganath %E Byron Wallace %E Jenna Wiens %F pmlr-v106-prabhu19a %I PMLR %J Proceedings of Machine Learning Research %P 532--552 %U http://proceedings.mlr.press %V 106 %W PMLR %X We consider the problem of clinical image classification for the purpose of aiding doctors in dermatological disease diagnosis. Diagnosis of dermatological conditions from images poses two major challenges for standard off-the-shelf techniques: First, the distribution of real-world dermatological datasets is typically long-tailed. Second, intra-class variability is large. To address the first issue, we formulate the problem as low-shot learning, where once deployed, a base classifier must rapidly generalize to diagnose novel conditions given very few labeled examples. To model intra-class variability effectively, we propose Prototypical Clustering Networks (PCN), an extension to Prototypical Networks (Snell et al. , 2017 ) that learns a mixture of “prototypes” for each class. Prototypes are initialized for each class via clustering and refined via an online update scheme. Classification is performed by measuring similarity to a weighted combination of prototypes within a class, where the weights are the inferred cluster responsibilities. We demonstrate the strengths of our approach in effective diagnosis on a realistic dataset of dermatological conditions.
APA
Prabhu, V., Kannan, A., Ravuri, M., Chaplain, M., Sontag, D. & Amatriain, X.. (2019). Few-Shot Learning for Dermatological Disease Diagnosis. Proceedings of the 4th Machine Learning for Healthcare Conference, in PMLR 106:532-552

Related Material