Improving Medication Sachet Identification in the Medido Medicine Clock
J.Q. Luik (TU Delft - Industrial Design Engineering)
C.A. Bakker – Mentor (TU Delft - Industrial Design Engineering)
M. Verwaal – Graduation committee member (TU Delft - Industrial Design Engineering)
S. Vermeulen – Graduation committee member (Evondos Group)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Medication sachet identification plays an important role in the safe and reliable operation of the Medido Medicine Clock, an automated medication sachet dispenser used by more than 20000 clients in home-care and intramural settings. Accurate identification supports medicine roll validation, enables advanced dispensing functionality such as Advanced Roll Loading, and provides care organizations with confidence that the intended medication sachet is dispensed. However, barcode reading performance in deployed Medido model M302 devices has proven inconsistent due to optical conditions, variation in medication sachet layouts, contamination of the optical path, and hardware configuration.
This thesis investigated the factors limiting barcode reading performance and translated these findings into an integrated improvement concept for the existing M302 platform. An exploratory experimental research approach was used, combining controlled barcode performance testing with analysis of barcode scanner sensor output. The influence of medicine roll transport conditions, illumination, mirror contamination, scanner positioning, and scanner behavior was systematically investigated using both software and hardware modifications.
The research demonstrated that barcode reading performance is affected by multiple interacting factors. Scanning during motion, optical contamination of the mirror, scanner positioning tolerances, limited grayscale dynamic range, and specular reflections from medication sachet materials all reduce decoding reliability. In particular, stationary barcode reading and alternative illumination strategies substantially improved performance. The findings showed that barcode reading performance cannot be addressed through a single intervention, but requires a combined optical, mechanical, and firmware-based solution.
Based on these insights, a final design was developed that combines backlighting through a revised Ultrasound Transmitter & Lighting PCBA, a Scan Engine LED Blocker, a Dust Cover, stationary barcode reading, and firmware-based reconstruction methods for incomplete scans.
Validation showed successful decode rates of 94% for top-positioned barcodes and 100% for bottom-positioned barcodes, resulting in an overall decode rate of 96%. The concept can be installed during refurbishment in approximately 15 minutes, at an estimated implementation cost of €25 per refurbished device. With a proposed development and validation trajectory of 6 to 9 months, the concept provides a feasible basis for improving medication sachet identification across refurbished M302 devices without replacing the installed fleet.