AITC Consortium | Basic Information | PennAITech Awardees | All Resources

Designing Usable Technologies for Older Adults via Data-Driven Whole-Person User Personas

by

Using machine learning to create data-driven user personas for the design of usable technologies for older adults
Awardee Organization(s): University of Minnesota

Principal Investigator(s): Robin Austin, PhD, DNP, DC, RN-BC
Official Project Title: Designing Usable Technologies for Older Adults via Data-Driven Whole-Person User Personas
AITC Partner: PennAITech
Website(s): www.nursing.umn.edu

The long-term goal of this research led by Dr. Robin Austin at the University of Minnesota and her team is to improve health outcomes by combining whole-person patient-generated health data with EHR data, to inform clinical conversations, predict patient trajectories, and identify appropriate interventions. This research, Designing Usable Technologies for Older Adults via Data-Driven Whole-Person User Personas, will create a set of data-driven user personas based on data from 6 studies where 783 adults 65+ years old independently completed a comprehensive health assessment using the MyStrengths+MyHealth (MSMH) mobile app developed by Dr. Robin’s team. MSMH assesses 42 strength/problem areas (e.g., Income, Spirituality, Nutrition) divided into four categories (e.g., My Living, My Mind and Networks, My Body, My Self- Care). Individuals can specify any of 335 challenges (e.g., Hard to concentrate) and any of 4 needs related to each strength/problem area (e.g., Check-ins, Hands-on Care, Info/Guidance, Care Coordination). We will use machine learning approaches, clustering analysis and association rule learning, which are frequently used to develop user personas. These whole-person user personas will account for a 360 degree view of the person, meaning the environments in which individuals live, their psychosocial and physical health needs, and their strengths. This research will inform person-centered technology design and develop a better understanding of the types of older adults who may use AI-based technologies.

View Resource
AITC Consortium | Basic Information | PennAITech Awardees | All Resources

Detecting Respiratory Distress in Patients with Advanced AD/ADRD Using Radio Sensors

by

AI-enabled near-field coherent sensing (NCS) radio sensors to detect respiratory distress for advanced AD/ADRD patients who are unable to self-report
Awardee Organization(s): Weill Cornell Medicine
Principal Investigator(s): Veerawat Phongtankuel, MD, MS
Official Project Title: Detecting Respiratory Distress in Patients with Advanced ADRD Using Radio Sensors
AITC Partner: PennAITech
Website(s): www.weill.cornell.edu

This project focuses on designing and developing conversational care technologies for older adults and their caregivers. In prior work, we surveyed informal caregivers and older adult care receivers to understand their care routines. Survey findings showed how nearly 20% of care partners used voice assistants in their homes, signaling an opportunity to extend research on older adults’ conversational technology use to include care partners. Next, we conducted a diary study and interviews with caregivers and care receivers to investigate gaps in care interactions and conversations. We found that care receivers experienced more communication frustrations than caregivers and that older adult caregivers wanted more opportunities to influence their care routines.
In this project, we will use these findings to develop in-home conversational technologies that use prompts to structure care conversations between older adults and their caregivers. We contribute a nuanced dyadic perspective to care relationships as most care research focuses solely on caregiver perspectives. We also extend conversational technology research beyond information seeking to include more social uses by developing conversational technology applications with mainstream voice technologies (e.g., Amazon Alexa) to support improved care relationships, social and emotional well-being, and quality of life.

View Resource
Benchmarking | Data | All Resources

Diverse and Generative ML benchmark (DIGEN)

by

A modern machine learning benchmark, which includes: 40 datasets in tabular numeric format specially designed to differentiate the performance of some of the leading Machine Learning (ML) methods, and a package to perform reproducible benchmarking that simplifies comparison of performance of the methods. DIGEN provides comprehensive information on the datasets, including: ground truth – a mathematical formula presenting how the target was generated for each of the datasets, the results of exploratory analysis, which includes feature correlation and histogram showing how binary endpoint was calculated, multiple statistics on the datasets, including the AUROC, AUPRC and F1 scores, each dataset comes with Receiver-Operating Characteristics (ROC) and Precision-Recall (PRC) charts for tuned ML methods, and a boxplot with projected performance of the leading methods after hyper-parameter tuning (100 runs of each method started with different random seed), Apart from providing a collection of datasets and tuned ML methods, DIGEN provides tools to easily tune and optimize parameters of any novel ML method, as well as visualize its performance in comparison with the leading ones. DIGEN also offers tools for reproducibility.

 

View Resource
Software | Machine Learning Modeling | Technology: Tools, Hardware, and Software | All Resources

Extended Supervised Tracking and Classification System (scikit-ExSTraCS)

by

The scikit-ExSTraCS package includes a sklearn-compatible Python implementation of ExSTraCS 2.0. ExSTraCS 2.0, or Extended Supervised Tracking and Classifying System, implements the core components of a Michigan-Style Learning Classifier System (where the system’s genetic algorithm operates on a rule level, evolving a population of rules with each their own parameters) in an easy to understand way, while still being highly functional in solving ML problems. It allows the incorporation of expert knowledge in the form of attribute weights, attribute tracking, rule compaction, and a rule specificity limit, that makes it particularly adept at solving highly complex problems. In general, Learning Classifier Systems (LCSs) are a classification of Rule Based Machine Learning Algorithms that have been shown to perform well on problems involving high amounts of heterogeneity and epistasis. Well designed LCSs are also highly human interpretable. LCS variants have been shown to adeptly handle supervised and reinforced, classification and regression, online and offline learning problems, as well as missing or unbalanced data. These characteristics of versatility and interpretability give LCSs a wide range of potential applications, notably those in biomedicine.

 

View Resource
AITC Consortium | Basic Information | PennAITech Awardees | All Resources

Feasibility of Digital Monitoring to Detect Autonomic Markers of Empathy Loss in bvFTD

by

Wearable sensors capturing digital autonomic biomarkers to detect empathy loss for frontotemporal dementia (bvFTD)
Awardee Organization(s): University of Pennsylvania Frontotemporal Degeneration Center
Principal Investigator(s): Emma Rhodes, PhD
Official Project Title: Feasibility of Digital Monitoring to Detect Autonomic Markers of Empathy Loss in bvFTD
AITC Partner: PennAITech
Website(s): www.med.upenn.edu/ftd
Loss of empathy is a core symptom of behavioral variant frontotemporal dementia (bvFTD) that negatively impacts daily functioning and is highly distressing to families and caregivers. Scientific research has struggled to understand the specific causes of empathy loss in bvFTD. A relatively unexplored but promising avenue of scientific inquiry is the role of autonomic nervous system (ANS) arousal in empathy loss in bvFTD. The ANS is comprised of two complementary subsystems, the sympathetic and parasympathetic nervous systems, which operate together to regulate an individual’s level of physiologic arousal in response to emotional cues from the environment. Patients with bvFTD show abnormalities in autonomic arousal that are linked to symptoms of social dysfunction, including loss of empathy, but this line of research has been hindered by reliance on traditional methods of measuring autonomic arousal, namely hard-wired EKG and skin conductance sensors, which restrict the movement of the patient and are sensitive to motion effects. Recent advances in wearable smartwatch technology allow for precise, unobtrusive measurement of autonomic arousal with built-in motion sensors that more accurately capture key arousal variables, such as respiratory sinus arrythmia and skin conductance. Use of smartwatch technology will advance our understanding of physiologic mechanisms of empathy loss in bvFTD and other neuropsychiatric symptoms in ADRD and identify potential treatment targets. The overarching goal of this project is to test the feasibility of using a smartwatch to capture abnormalities in autonomic arousal in bvFTD and validate digital markers of autonomic abnormalities against behavioral measures of empathy loss.
In this project, we will use these findings to develop in-home conversational technologies that use prompts to structure care conversations between older adults and their caregivers.
We contribute a nuanced dyadic perspective to care relationships as most care research focuses solely on caregiver perspectives. We also extend conversational technology research beyond information seeking to include more social uses by developing conversational technology applications with mainstream voice technologies (e.g., Amazon Alexa) to support improved care relationships, social and emotional well-being, and quality of life.

View Resource
Internal | Artificial Intelligence | Training Resources | All Resources

Foundations of Artificial Intelligence

by

Educational lectures for the course: “Foundations of Artificial Intelligence” developed by Dr. Ryan Urbanowicz in 2020 at the University of Pennsylvania’s Perelman School of Medicine. This is the first of three courses covering topics in artificial intelligence for application within the context of informatics and biomedical research. The course is divided into modules that cover (1) introductory/background materials, (2) logic, (3) other knowledge representation, (4) essentials of expert systems, (5) search, (6) uncertainty, and (7) advanced/auxiliary topics. These topics offer a global foundation for branches of AI application and research, including concepts that will later support a deeper understanding of inductive reasoning and machine learning. In a practical sense, this course focuses on how biomedical data can be organized, represented, interpreted, searched, and applied in order to derive knowledge, make decisions, and ultimately make predictions while avoiding bias. This course was assembled using content from a wide variety of textbooks, slides, and lectures by various authors and speakers on the relevant topics. Some lectures were prepared and given by guest lecturers and thus have not been posted. At the time of posting, this course is in its second year so any feedback is welcome regarding any mistakes or suggested improvements.

View Resource
Clinical/biomedical | Penn Resources | All Resources

Frontotemporal Degeneration Center (FTDC)

by

The Penn Frontotemporal Degeneration Center brings together an energetic team of creative clinicians and researchers dedicated to the investigation and treatment of early onset neurodegenerative conditions. The research expertise at the Penn FTD Center spans many levels of neuroscience ranging from detailed clinico-pathological studies, biomarker discovery, genetics, neuropsychological studies, functional and structural neuroimaging, and cognitive neuroscience investigations of language, memory, and social cognition.

View Resource
Software | Data simulation | Technology: Tools, Hardware, and Software | All Resources

Genetic Architecture Model Emulator for Testing and Evaluating Software (GAMETES)

by

GAMETES is an algorithm for the generation of complex single nucleotide polymorphism (SNP) models for simulated association studies. GAMETES is designed to generate epistatic models which we refer to as pure and strict. These models constitute the worst-case in terms of detecting disease associations, since such associations may only be observed if all n loci are included in the disease model. The user-friendly GAMETES software rapidly and precisely generates epistatic multi-locus models, and using these models, can also generate simulated datasets exhibiting epistasis. Version 2.2 adds the ability to generate heterogeneous datasets by applying multiple independent models to different subsets of the simulated data. Further additional features include the facility to create additive datasets by applying multiple independent models to the entire dataset, as well as functionality for the design of continuous endpoints. Additionally, we have added a custom model generation feature, so that users may directly specify and examine the properties of any 2 or 3 locus SNP model. Simple Mendelian models may also be generated with this feature.

 

View Resource