Branding
Hospital Records Digitization Agent
An AI agent built for LaserEye Centre in Westlands, Nairobi — processing uploaded medical record images, extracting patient data using AI vision, and storing it in a structured database automatically.
Year :
2026
Industry :
Healthcare
Client :
LaserEye Centre
Project Duration :
3 weeks
Problem :
LaserEye Centre, a specialist eye clinic in Westlands, Nairobi, was managing a high volume of patient records in physical and scanned image form. Retrieving and entering records was slow, error-prone, and entirely dependent on manual data entry — consuming staff time and introducing risk of data loss or inaccuracy.
Solution :
I built an AI agent on n8n that receives uploaded medical record images via a simple interface. The agent passes each image to an AI vision model that performs OCR and data extraction — identifying patient name, date of birth, diagnosis, prescription, and attending doctor — then structures the data and writes it directly into the clinic's database. No manual typing required.
Challenge :
Medical data requires a high accuracy threshold — guessing is not acceptable. The main challenge was building reliable error handling: if the agent couldn't confidently extract a field, it needed to flag the record for human review rather than fill it with incorrect data. Balancing automation speed with medical-grade accuracy was the core technical challenge.
Summary :
The Hospital Records Digitization Agent reduced manual data entry time per record from several minutes to near zero, standardised the data structure, and made records instantly searchable. It's a real deployment running for a real client — and a demonstration of how AI agents can modernise operations in sectors that still run on paper.
More Projects
Branding
Hospital Records Digitization Agent
An AI agent built for LaserEye Centre in Westlands, Nairobi — processing uploaded medical record images, extracting patient data using AI vision, and storing it in a structured database automatically.
Year :
2026
Industry :
Healthcare
Client :
LaserEye Centre
Project Duration :
3 weeks
Problem :
LaserEye Centre, a specialist eye clinic in Westlands, Nairobi, was managing a high volume of patient records in physical and scanned image form. Retrieving and entering records was slow, error-prone, and entirely dependent on manual data entry — consuming staff time and introducing risk of data loss or inaccuracy.
Solution :
I built an AI agent on n8n that receives uploaded medical record images via a simple interface. The agent passes each image to an AI vision model that performs OCR and data extraction — identifying patient name, date of birth, diagnosis, prescription, and attending doctor — then structures the data and writes it directly into the clinic's database. No manual typing required.
Challenge :
Medical data requires a high accuracy threshold — guessing is not acceptable. The main challenge was building reliable error handling: if the agent couldn't confidently extract a field, it needed to flag the record for human review rather than fill it with incorrect data. Balancing automation speed with medical-grade accuracy was the core technical challenge.
Summary :
The Hospital Records Digitization Agent reduced manual data entry time per record from several minutes to near zero, standardised the data structure, and made records instantly searchable. It's a real deployment running for a real client — and a demonstration of how AI agents can modernise operations in sectors that still run on paper.
More Projects
Branding
Hospital Records Digitization Agent
An AI agent built for LaserEye Centre in Westlands, Nairobi — processing uploaded medical record images, extracting patient data using AI vision, and storing it in a structured database automatically.
Year :
2026
Industry :
Healthcare
Client :
LaserEye Centre
Project Duration :
3 weeks
Problem :
LaserEye Centre, a specialist eye clinic in Westlands, Nairobi, was managing a high volume of patient records in physical and scanned image form. Retrieving and entering records was slow, error-prone, and entirely dependent on manual data entry — consuming staff time and introducing risk of data loss or inaccuracy.
Solution :
I built an AI agent on n8n that receives uploaded medical record images via a simple interface. The agent passes each image to an AI vision model that performs OCR and data extraction — identifying patient name, date of birth, diagnosis, prescription, and attending doctor — then structures the data and writes it directly into the clinic's database. No manual typing required.
Challenge :
Medical data requires a high accuracy threshold — guessing is not acceptable. The main challenge was building reliable error handling: if the agent couldn't confidently extract a field, it needed to flag the record for human review rather than fill it with incorrect data. Balancing automation speed with medical-grade accuracy was the core technical challenge.
Summary :
The Hospital Records Digitization Agent reduced manual data entry time per record from several minutes to near zero, standardised the data structure, and made records instantly searchable. It's a real deployment running for a real client — and a demonstration of how AI agents can modernise operations in sectors that still run on paper.