AI in Healthcare

AI in Healthcare further empowers users to play a critical role by enhancing clinical decision-making with machine learning to build the ‘treatments of the future’. It helps practitioners learn to build, evaluate, and integrate predictive models that have the power to transform patient outcomes. The main approach is by classifying and segmenting 2D and 3D medical images to augment diagnosis and then move on to modeling patient outcomes with electronic health records to optimize clinical trial testing decisions. In your IDE build Python algorithms collect data from wearable devices to estimate the wearer’s pulse rate in the presence of motion.

Increasing Healthcare Team Capabilityby Task

  • Recommend appropriate imaging modalities for common clinical applications of 2D medical imaging
  • Perform exploratory data analysis (EDA) on 2D medical imaging data to inform model training and
    explain model performance
  • Establish the appropriate ‘ground truth’ methodologies for training algorithms to label medical images
  • Extract images from a DICOM dataset
  • Train common CNN architectures to classify 2D medical images

Increasing Healthcare Team Capability-by Functional Area

Applying AI to EHR Data

With the transition to electronic health records (EHR) over the last decade, the amount of EHR data has increased exponentially, providing an incredible opportunity to unlock this data with AI to benefit the healthcare system. Enhance the fundamental skills of working with EHR data in order to build and evaluate compliant, interpretable machine learning models that account for bias and uncertainty using cutting-edge libraries and tools including Tensorflow Probability, Aequitas, and Shapley.

Understand the implications of key data privacy and security standards in healthcare.

AI can help apply industry code sets (ICD10-CM, CPT, HCPCS, NDC), transform datasets at different
EHR data levels, and use Tensorflow to engineer features.

2D Medical Imaging Data

2D imaging, such as X-ray, is widely used when making critical decisions about patient care and accessible by most healthcare center s around the world. With the advent of deep learning for non-medical imaging data over the past half decade, the world has quickly turned its attention to how AI could be specifically applied to medical imaging to improve clinical decision-making and to optimize workflows. Learn the fundamental skills needed to work with 2D medical imaging data and how to use AI to derive clinically-relevant insights from data gathered via different types of 2D medical imaging such as x-ray, mammography, and digital pathology.
Extract 2D images from DICOM files and apply the appropriate tools to perform exploratory data analysis on them. Build different AI models for different clinical scenarios that involve 2D images and learn how to position AI tools for regulatory approval.

3D Medical Imaging Data

3D medical imaging exams such as CT and MRI serve as critical decision-making tools in the clinician’s everyday diagnostic armamentarium. These modalities provide a detailed view of the patient’s anatomy and potential diseases, and are a challenging though highly promising data type for AI applications. The fundamental skills needed to work with 3D medical imaging datasets and frame insights derived from the data in a clinically relevant context. These images are acquired, stored in clinical archives, and subsequently read and analyzed. Clinicians use 3D medical images in practice and where AI holds most potential in their work with these images. Our platforms help design and apply machine learning algorithms to solve the challenging problems in 3D medical imaging and how to integrate the algorithms into the clinical workflow.

Applying AI to Wearable Device Data

Wearable devices are an emerging source of physical health data. With continuous, unobtrusive monitoring they hold the promise to add richness to a patient’s health information in remarkable ways. Understand the functional mechanisms of three sensors (IMU, PPG, and ECG) that are common to most wearable devices and the foundational signal processing knowledge critical for success in this domain. Attribute physiology and environmental context’s effect on the sensor signal. Build algorithms that process the data collected by multiple sensor streams from wearable devices to surface insights about the wearer’s health.