Explore My Work

Here you can find the technical details of my past and current projects, including their outcomes and the tools I used to get the job done.Get in touch today for a consultation.

Ditti Research Dashboard

2022 – OngoingUPenn Geriatric Sleep Research Lab
Try a live demo
The Ditti Research Dashboard was created out of the need for a single, centralized platform for managing and visualizing data collected from the Penn Ditti Mobile App. With the Ditti Research Dashboard, coordinators and clinicians can quickly enroll new subjects and monitor their adherence to behavioral sleep interventions with ease.
Dashboard for Research and Clinical Data ManagementThe Ditti Research Dashboard allows researchers and clinicians to manage and visualize anonymized study data from the Penn Ditti app for monitoring behavioral sleep intervention adherence.
Comprehensive Role and Permission ControlThe Admin Dashboard provides granular controls for research coordinator roles, enabling specific access management across studies and app-wide permissions.
Scalable and HIPAA-compliant Cloud InfrastructureBuilt on AWS with TypeScript, React.js, and Python, the app integrates user data from DynamoDB, and is expanding to include a FitBit Dashboard for visualizing participant sleep data.
Key Features
For Researchers
  • Visualizations of user interactions with the Penn Ditti Mobile App
  • Interfaces for managing study-related data and enrolling study subjects
  • Tools for labeling and uploading audio files for the Penn Ditti Mobile App
  • Administrative controls for managing coordinator-level and study-level permissions
  • Serverless architecture for controlling costs on an on demand basis
Upcoming
  • Integrations with third-party wearable device APIs
  • Side-by-side visualizations of wearable and Penn Ditti data
Tech Stack & Infrastructure
Backend
  • Python
  • Flask Web Framework
  • PostgreSQL
Frontend
  • TypeScript
  • React.js
  • Tailwind CSS
  • Visx
AWS
  • Lambda
  • Cognito
  • AppSync
  • DynamoDB
  • S3
  • CloudFront
  • Secrets Manager
  • Relational Database Service
Related Publications
McPhillips, M. V., Li, J., Petrovsky, D. V., Brewster, G. S., Ward, E. J., 3rd, Hodgson, N., & Gooneratne, N. S. (2023). Assisted Relaxation Therapy for Insomnia in Older Adults With Mild Cognitive Impairment: A Pilot Study. International journal of aging & human development, 97(1), 65–80. https://doi.org/10.1177/00914150221132163McPhillips, M. V., Li, J., Petrovsky, D. V., Gooneratne, N. S., Aryal, S., & Hodgson, N. A. (2023). A randomized controlled trial to test a behavioral sleep intervention to improve insomnia symptoms in older adults with mild cognitive impairment: Multicomponent Behavioral Sleep Intervention (MBSI) protocol. Contemporary clinical trials, 127, 107137. https://doi.org/10.1016/j.cct.2023.107137

Penn Ditti Mobile Application

2024 – OngoingUPenn Geriatric Sleep Research Lab
The Penn Ditti Mobile Application is a tool that researchers and clinicians in sleep medicine can use for relaxation and mindfulness exercises for improving sleep. It fills a gap in monitoring adherence to behavioral sleep interventions by enabling clinicians and researchers to monitor when and how users are interacting with the app.
Behavioral Sleep Intervention and Data TrackingPenn Ditti provides a breathing exercise feature where users synchronize breaths with screen taps to enhance relaxation and sleep quality; each tap is timestamped and stored in AWS DynamoDB to monitor adherence.
Sleep Hygiene and Audio for MindfulnessThe app includes a Sleep Hygiene Information page with evidence-based sleep tips, and a new audio feature allowing users to play short mindfulness audio tracks, with interactions recorded for clinical tracking.
Enhanced Data Management and ReportingAutomated data archiving and email reporting ensure secure, accessible clinical data storage, supporting app reliability and compliance with clinical standards.
Key Features
For Users
  • Guidance on maintaining healthy sleep hygiene
  • Somatic mindfulness exercises for better sleep
  • Calming audio files for relaxation and meditation
Upcoming
  • Activity summaries for tracking changes in sleep patterns
Tech Stack & Infrastructure
Backend
  • AWS Amplify
  • GraphQL
Mobile
  • React Native
AWS
  • Amplify
  • AppSync
  • DynamoDB
  • Cognito
Related Publications
McPhillips, M. V., Li, J., Petrovsky, D. V., Brewster, G. S., Ward, E. J., 3rd, Hodgson, N., & Gooneratne, N. S. (2023). Assisted Relaxation Therapy for Insomnia in Older Adults With Mild Cognitive Impairment: A Pilot Study. International journal of aging & human development, 97(1), 65–80. https://doi.org/10.1177/00914150221132163McPhillips, M. V., Li, J., Petrovsky, D. V., Gooneratne, N. S., Aryal, S., & Hodgson, N. A. (2023). A randomized controlled trial to test a behavioral sleep intervention to improve insomnia symptoms in older adults with mild cognitive impairment: Multicomponent Behavioral Sleep Intervention (MBSI) protocol. Contemporary clinical trials, 127, 107137. https://doi.org/10.1016/j.cct.2023.107137

BOOST-3 & PRECICECAP Data Portals

2020 – 2022Moberg Analytics, Inc.
The Brain Oxygen Optimization in Severe TBI Phase-3 (BOOST-3) and PREcision Care in Cardiac arrest - ICECAP ancillary are two clinical trials that seek to enhance care for patients with traumatic brain injury (TBI). The data portals developed for these studies support novel insights by researchers by managing study-wide data in a scalable manner.
Scalable Management of Multi-Terabyte DataIBM Object Storage is used to store multiple terabytes of high-resolution time-series physiological data while enabling rapid access across study sites in Canada and the U.S.
Leverage of Cloud-Native HDF5 Format For Rapid AnalysisData is converted from a binary format recorded by a medical device to the Hierarchical Data Format 5 (HDF5), enabling rapid analysis on cloud-based distributed systems.
Standardized Labeling for Tracking Study-Wide TrendsAn extensible PostgreSQL database schema enables linking datasets with descriptive metadata like patient characteristics and study site and mapping of data sources to standardized labels.
Key Features
For Researchers
  • Interfaces for uploading and labeling participant data
  • Live tracking of asynchronous data processing tasks
  • Summaries of available modalities of uploaded data
Tech Stack & Infrastructure
Backend
  • Python
  • Flask Web Framework
  • PostgreSQL
Frontend
  • HTML/CSS
  • JavaScript
IBM Cloud
  • Kubernetes
  • Object Storage
Related Publications
Foreman, B., Lissak, I. A., Kamireddi, N., Moberg, D., & Rosenthal, E. S. (2021). Challenges and Opportunities in Multimodal Monitoring and Data Analytics in Traumatic Brain Injury. Current neurology and neuroscience reports, 21(3), 6. https://doi.org/10.1007/s11910-021-01098-yElmer, J., He, Z., May, T., Osborn, E., Moberg, R., Kemp, S., Stover, J., Moyer, E., Geocadin, R. G., Hirsch, K. G., & PRECICECAP Study Team (2022). Precision Care in Cardiac Arrest: ICECAP (PRECICECAP) Study Protocol and Informatics Approach. Neurocritical care, 37(Suppl 2), 237–247. https://doi.org/10.1007/s12028-022-01464-9

Repositories

bert-small-ko-word-boundary-detection
Python 69.6%
Jupyter Notebook 30.4%
MIT LicenseDecember 3, 2024

Get in Touch

Do you have an idea for an app or have an upcoming data-driven study? Please provide a brief introduction and description of your project below and let’s get in touch.
0 / 1,000