WizTherapy, a pioneer in neurodivergent client care, is known for blending empathy with innovation. But behind the scenes, their clinicians and staff were burdened with time-consuming paperwork, especially during the intake process. As patient volume grew, so did the inefficiencies in managing diverse document formats and manually converting them into structured data.  

Recognizing the need for speed, accuracy, and compliance, WizTherapy decided to invest in clinical workflow automation to streamline provider operations and enhance the client experience.

The problem: operational inefficiency from manual intake workflows

Every new client requires a completed intake form, often submitted as PDFs, Word docs, or scanned images. Clinicians had to manually extract information, standardize it, and re-enter it into digital systems.

This slowed down onboarding, increased the risk of data entry errors, and created friction in the client experience. Without structured schema outputs, integration with their React JSON Schema Form (RJSF)-based systems remained limited. The need of the hour: a smart clinical workflow automation solution that can handle unstructured documents and convert them into structured, editable schemas.

The solution: AI-driven clinical workflow automation with JSON schema generation

GoML delivered a 4-week PoC purpose built for clinical workflow automation. The solution enabled document ingestion, AI-powered data extraction, and automatic schema creation, all while maintaining transparency, security, and ease of use.

Multi-format intake ingestion and preprocessing

The solution allowed clinical teams to upload intake forms in PDF, Word, or image format via a secure web UI or API. AWS-native services, including Lambda and Textract, were used for processing.

LLM-based entity extraction and schema creation

Claude 3.5 on Amazon Bedrock identified and extracted form fields using AI/NLP techniques. It then generated structured JSON and UI schemas compatible with RJSF, reducing manual mapping and enabling dynamic form rendering.

Interactive schema preview and editing

A lightweight ReactJS interface lets users

• Upload and preview extracted schema

• Make minor changes to labels, field types, or validation rules

• Download the final schema for integration

Compliance-first design with monitoring and encryption

With clinical workflow automation, HIPAA compliance was top-of-mind. All extracted data was logged securely, encrypted using AWS KMS, and monitored through CloudWatch. Discrepancies or incomplete data were flagged for human review.

The impact: intelligent automation across clinical operations

• 80% reduction in manual effort related to intake form creation and digitization

• 90% accuracy in schema generation, even across varied document formats

• Accelerated clinical onboarding, enabling faster care delivery and reduced administrative load

Lessons for digital health and clinical teams

Common pitfalls to avoid

• Underestimating the variability of document formats

• Relying solely on manual schema conversion

• Skipping audit trails in sensitive clinical workflows

Tips for growth-oriented care platforms

• Treat intake forms as structured data assets

• Integrate AI to accelerate both admin and care workflows

• Use RJSF for dynamic, low-code intake interfaces backed by JSON schemas

Ready to simplify your clinical workflows with AI?

GoML’s clinical workflow automation platform can transform how your care team handles intake, documentation, and schema management so they can focus more on care and less on complexity.

→ Contact us to build your first AI-powered clinical automation prototype.