Development and Validation of a Digital Image Processing-Based Pill Detection Tool for an Oral Medication Self-Monitoring System

Autores da FMUP

• Rita Da Silva Amaral

  Autor

• Rute Alexandra Borges De Almeida

  Autor

• Cristina Isabel Oliveira Jácome

  Autor

• Ana Margarida Barbosa Ribeiro Pereira

  Autor

• João De Almeida Lopes Da Fonseca

  Autor

Participantes de fora da FMUP

• Holtk?tter, J
• Cardoso, R
• Pereira, A.
• Chon, KH

Unidades de investigação

• Laboratório Associado RISE- Rede de Investigação em Saúde

  

  Investigador

  Fernando Carlos De Landér Schmitt

• Medicina da Comunidade, Informação e Decisão em Saúde

• RISE-Health

  

  Investigador

  Fernando Carlos De Landér Schmitt

Abstract

Long-term adherence to medication is of critical importance for the successful management of chronic diseases. Objective tools to track oral medication adherence are either lacking, expensive, difficult to access, or require additional equipment. To improve medication adherence, cheap and easily accessible objective tools able to track compliance levels are necessary. A tool to monitor pill intake that can be implemented in mobile health solutions without the need for additional devices was developed. We propose a pill intake detection tool that uses digital image processing to analyze images of a blister to detect the presence of pills. The tool uses the Circular Hough Transform as a feature extraction technique and is therefore primarily useful for the detection of pills with a round shape. This pill detection tool is composed of two steps. First, the registration of a full blister and storing of reference values in a local database. Second, the detection and classification of taken and remaining pills in similar blisters, to determine the actual number of untaken pills. In the registration of round pills in full blisters, 100% of pills in gray blisters or blisters with a transparent cover were successfully detected. In the counting of untaken pills in partially opened blisters, 95.2% of remaining and 95.1% of taken pills were detected in gray blisters, while 88.2% of remaining and 80.8% of taken pills were detected in blisters with a transparent cover. The proposed tool provides promising results for the detection of round pills. However, the classification of taken and remaining pills needs to be further improved, in particular for the detection of pills with non-oval shapes.

Keywords

• computer vision; image processing; medication adherence; object detection; pill detection