An Advanced Electronic Health Record for Prescription Analysis Using YOLO and OCR with Spell Correction

Keeping accurate and well-organized medical records is still quite difficult in today's healthcare setting, particularly when it comes to handwritten prescriptions. In order to solve this problem, RxVision created a sophisticated technology that automatically transforms handwritten prescriptions into organized Electronic Health Records (EHRs). The system uses OCR for text extraction following utilizing the YOLO model to identify important sections of interest, such as medication names, dosing amounts, and usage instructions. A spell correction module improves OCR outputs to increase accuracy and reduce handwritten text interpretation problems. With a mAP@0.5 of 0.93, Precision of 94.1%, Recall of 91.3%, and an F1-score of 0.92, the final integrated model demonstrated its dependability and effectiveness with outstanding performance. Furthermore, the structured prescription data generated by the system enables downstream analytical tasks such as risk pattern identification and trend analysis, supporting clinicians in data-driven decision-making. All things considered, RxVision greatly enhances healthcare data management by making patient record maintenance and utilization quicker, more accurate, and simpler for practitioners and hospitals. This system way more managing the health records easier, more accurate, and faster, reducing the manual work for the hospitals and helping the doctors to make better and data-driven decisions