Artificial Intelligence (AI) refers to a set of technologies that allow machines and computers to simulate human intelligence. AI technologies have been developed to analyze a diverse array of health data, including patient, behavioral, environmental, clinical, and drug data. AI technologies have also been used extensively in analyzing clinical data, including medical images, electronic health records (EHRs), and physiological signals. Wide deployment of consumer grade devices and continuous monitoring of physiological data can be a low-cost solution for personal healthcare. However, healthcare data acquired from these sensors cannot be manually analyzed due to its scale and therefore automated data analytics and decision support techniques must be developed.

Challenges remain regarding the use of AI in medical images and biosignal analysis. Analysis of medical images relies heavily on deep learning architectures that were designed and trained on natural images. These models require very large amounts of data to train. However, in case of rare diseases, usually there is little data for training available. The analysis of physiological signals is computationally very expensive and highly influenced by noise and measurement errors. Developing noise-resistant AI approaches is important for analysis physiological data acquired by lower quality consumer-grade devices. Conventional machine learning models for analyzing structured information in EHRs are mostly vector based, composed of the summary statistics of the values of the features in different dimensions. The analysis is often hindered by missing data, which requires data imputation if analyzed.

Linking patient and behavioral data with other types of clinical and environment data can provide a unique opportunity to impact healthcare. The development of new integrative AI-based data analytics methods that can analyze heterogeneous data from sources, such as sensor, keyboard, voice, and speech data from low quality sensors (such as obtained by consumer smartphones and wristbands) is a highly promising research direction. Specifically, data-driven AI approaches to analyze patient behaviors captured by surrounding smart devices, such as installed in smart homes or from wearable sensors or Intelligent Internet of Health Things, can contribute to preventive healthcare quality improvements.

We invite researchers to contribute original research articles as well as review or methodological articles that will stimulate continuing efforts to develop and improve AI methods for consumer healthcare applications.

Potential topics include but are not limited to the following:

• Artificial intelligence for preventive healthcare
• Analysis of low-quality data from early medical diagnostics
• Integrative analysis of medical data
• Temporal data mining in electronic health records
• Digital phenotyping
• Data mining methods for outlier detection in streaming medical data
• Machine learning from small, imperfect medical data
• Noise resistant machine learning methods
• Deep neural networks for understanding noisy sensor data
• Intelligent Internet of Health Things
• Artificial intelligence solutions for consumer health analytics
• Data analytics for healthcare-as-a-service
• Smartphone consumer applications for e-health
• Case studies of smart consumer innovation for healthcare