CALL FOR PAPERS
Applied Data Science for Healthcare Workshop in KDD 2023: Applications and New Frontiers of Generative Models for Healthcare
4 page submissions are due by May 23, 2023
https://dshealthkdd.github.io/dshealth-2023
https://kdd.org/kdd2023/workshops/
Generative models have a long history, and many application areas exist in medical machine learning (ML) and artificial intelligence (AI). In healthcare research, one of the most common applications of generative models has been generating synthetic data for training machine learning models. It is often used to increase the representation of patient subgroups to improve generalization and mitigate algorithmic biases. This is especially valuable in application domains where data is hard to come by. The generative models can also be used for specific model evaluation purposes (e.g., within a robustness or generalizability assessment, virtual clinical trials). They can help generate synthetic ground truth data when data labeling is highly burdensome. Moreover, generative models have been successfully applied in data preprocessing or enhancement, such as image reconstruction or denoising deep learning algorithms in medical imaging. While such generative models have proven their utility in the health domain, many open questions remain concerning the approaches for evaluating their effectiveness and safety. Testing and evaluation of such models require specific considerations. Taking the assessment of the gap between the generated data and the reality — the so-called Sim2Real challenge — as an example, it is often unclear how to (i) quantify this domain gap and its impact on downstream performance in a meaningful manner and (ii) reduce it to leverage the potential of generative models fully. New challenges are also emerging on a more grand scale. The recent advances in Large Language Models (LLMs) make data generation even more effortless. However, the misinformation generated with such models may cause a “pollution” of data for future model training. We can expect an increased need for effective fact-checking approaches. Despite the considerable growth of this area of research, the actual use of NLP technology for fact-checking is still in its infancy. In this half-day workshop, we would like to discuss some of the most common applications of generative models in ML/AI research in the healthcare domain, the current challenges, and also explore the potential new application areas.
We invite full papers and work-in-progress on the application of data science in healthcare. Topics may include but are not limited to the following topics (For more information, see workshop overview https://dshealthkdd.github.io/dshealth-2023/#home) with a special focus on generative models for healthcare.
- Synthetic data - Training data augmentation, e.g., in computer vision, medical imaging algorithm - Physics- and Chemistry- based generative models - Simulated data and privacy-preserving algorithms - In-silico clinical trials - Testing data, e.g., synthetic ground truth - Generative AI for tabular data - Interpretability - Privacy and security of generative AI - Inverse models for source verification - Watermark for AI-generated data - Factual capabilities of generative AI - Testing and evaluation of the generative models - Sim2Real domain gap - Data selection & quality aspects of the data (distribution shifts, monitoring of the models) - Fact-checking - Generating new healthcare-specific benchmarks - Bias detection and mitigation in healthcare - Reliability and trustworthiness of the generative models (actionable plans) - Application of LLMs - Systematic literature review - Modernizing pharmaceutical call center operations - Chatbot for patient registration, triage, scheduling, and rooming - Semantic data augmentation - Others - Responsible use of Generative AI - Generative AI Fairness and Bias detection - Generative AI bias mitigation (e.g., adversarial training) - Generative AI model transparency - Generative AI ethics and responsible AI risk management - Other - Knowledge representation learning
Papers must be submitted in PDF format to easychair https://easychair.org/conferences/?conf=dshealth2023 and formatted according to the new Standard ACM Conference Proceedings Template https://www.acm.org/publications/proceedings-template. Authors are encouraged to use the Overleaf template https://www.overleaf.com/latex/templates/acm-conference-proceedings-primary-article-template/wbvnghjbzwpc. Papers must be a maximum length of 4 pages, excluding references.
The program committee will select the papers based on originality, presentation, and technical quality for spotlight and/or poster presentation. Previous Iterations
- KDD Health Day - DSHealth 2022 https://dshealthkdd.github.io/dshealth-2022/: 2022 KDD Workshop on Applied Data Science for Healthcare: Transparent and Human-centered AI - KDD Health Day - DSHealth 2021 https://dshealthkdd.github.io/dshealth-2021/: Joint KDD 2021 Health Day and 2021 KDD Workshop on Applied Data Science for Healthcare State of XAI and Trustworthiness in Health - DSHealth 2020 https://dshealthkdd.github.io/dshealth-2020/: 2020 KDD Workshop on Applied Data Science for Healthcare: Trustable and Actionable AI for Healthcare - DSHealth 2019 https://dshealthkdd.github.io/dshealth-2019/: 2019 KDD Workshop on Applied Data Science for Healthcare: Bridging the Gap between Data and Knowledge - MLMH 2018 https://mlmhworkshop.github.io/mlmh-2018/: 2018 KDD Workshop on Machine Learning for Medicine and Healthcare
Follow us on Twitter: https://twitter.com/HealthDs https://dshealthkdd.github.io/dshealth-2023