Background

Tuberous Sclerosis Complex (TSC) is a rare genetic disorder that causes the development of tumors in various organs, primarily affecting the brain, eyes, heart, kidneys, skin, and lungs. The incidence of TSC is estimated to be approximately 1 case per 6,000–10,000 live births. Although TSC is a rare disease, it has brought misfortune to one million people worldwide.

TSC-associated neuropsychiatric disorders (TAND) encompasses various functional and clinical manifestations of brain dysfunction in TSC. These manifestations include aggressive behaviors, autism spectrum disorder (ASD), intellectual disabilities, psychiatric disorders, neuropsychological deficits, and difficulties in school and occupational settings. TAND seriously affects the mental health and well-being of TSC patients and their caregivers..

Significance and Goal

Extensive assessment data is crucial for rare diseases; accumulating large-scale patient data enables AI-driven analysis and prediction. In TAND, many patients experience epilepsy, particularly complex partial and refractory ones, often preceded by emotional fluctuations. With sufficient emotional data preceding seizures, deep learning models could be trained to predict the probability of impending seizure based on patients' TAND-related emotional changes.

The main goal of this project is to:

Discover: Is TAND's assessment a problem, and why?

To evaluate challenges in TAND assessment, I conducted a multi-perspective study involving literature review, analysis of existing tools, expert consultation on data availability, and patient interviews exploring user experiences and unmet needs.

Define: What Is the main problem I want to solve?

1. Research: Data Scarcity in TAND Studies through Self-Reporting
Limited TSC patient numbers hinder global TAND data collection, especially from adults and lower-income countries. Self-report methods can address this gap, supporting studies on epilepsy drug impacts on TAND.

2. Current Tools: Lack of a user-friendly and professional TAND self-report tool.
Official TAND tools (e.g., TAND-SQ) often offer poor user experience and limited feedback, while community-developed tools (e.g., dianriji) lack professional rigor and database synchronization.

3. Export Interview: Self-reported TAND data is currently lacking in the TSC Database.
To assess self-reported data collection, I contacted the TSC Natural History Database, a major source for TSC studies, regarding TAND data. They acknowledged issues like missing self-report data and limited scale granularity, and are actively working to address these concerns.

4. Patient Interview: Lack of clarity on the value of self-reporting, and it's not easy enough to use.
I contacted the Beijing Bowknot Tuberous Sclerosis Rare Disease Care Center and interviewed its president, who is also a patient, along with 4 other TAND patients from this association.

Ideate and prototype: How can I solve those problems?

User flow

Visual Design

I use the System Usability Scale (SUS) via in-app surveys to gather feedback from TAND patients, identify areas for improvement, and enhance the app's usability to better support their needs.