Collaborative Human-AI (CHAI): Evidence-based interpretable melanoma classification in dermoscopic images Conference Paper
| Authors: | Codella, N. C. F.; Lin, C. C.; Halpern, A.; Hind, M.; Feris, R.; Smith, J. R. |
| Title: | Collaborative Human-AI (CHAI): Evidence-based interpretable melanoma classification in dermoscopic images |
| Conference Title: | 1st International Workshop on Interpretability of Machine Intelligence in Medical Image Computing (iMIMIC) |
| Abstract: | Automated dermoscopic image analysis has witnessed rapid growth in diagnostic performance. Yet adoption faces resistance, in part, because no evidence is provided to support decisions. In this work, an approach for evidence-based classification is presented. A feature embedding is learned with CNNs, triplet-loss, and global average pooling, and used to classify via kNN search. Evidence is provided as both the discovered neighbors, as well as localized image regions most relevant to measuring distance between query and neighbors. To ensure that results are relevant in terms of both label accuracy and human visual similarity for any skill level, a novel hierarchical triplet logic is implemented to jointly learn an embedding according to disease labels and non-expert similarity. Results are improved over baselines trained on disease labels alone, as well as standard multiclass loss. Quantitative relevance of results, according to non-expert similarity, as well as localized image regions, are also significantly improved. © Springer Nature Switzerland AG 2018. |
| Keywords: | neuroimaging; melanoma; dermoscopy; oncology; image enhancement; artificial intelligence; diagnosis; medical computing; dermatology; evidence; computation theory; deep learning; computer aided instruction; image classification; explainable; global average pooling; interpretable; triplet-loss; weighted activation maps; nearest neighbor search; activation maps |
| Journal Title | Lecture Notes in Computer Science |
| Volume: | 11038 LNCS |
| Conference Dates: | 2018 Sept 16 |
| Conference Location: | Granada, Spain |
| ISBN: | 0302-9743 |
| Publisher: | Springer |
| Location: | Cham, Switzerland |
| Date Published: | 2018-01-01 |
| Start Page: | 97 |
| End Page: | 105 |
| Language: | English |
| DOI: | 10.1007/978-3-030-02628-8_11 |
| PROVIDER: | scopus |
| DOI/URL: | |
| Notes: | This was a half-day satellite event of the 21st International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) -- Chapter in "Understanding and Interpreting Machine Learning in Medical Image Computer Applications" (ISBN: 978-3-030-02627-1) -- Conference Paper -- Export Date: 3 December 2018 -- Source: Scopus |
