<![CDATA[Tuberculosis remains a major infectious disease worldwide, with medication non-adherence a critical barrier to successful treatment. Patients with visual impairments face additional challenges in medication management due to their inability to read medication labels, increasing the risk of medication errors and non-adherence. This systematic review analyzed 23 studies on the application of artificial intelligence (AI) to improve medication adherence in tuberculosis (TB) patients, with a particular focus on the potential of AI voice assistance for visually impaired patients. Results showed that AI such as machine learning (ML), natural language processing (NLP), and computer vision effectively improved the accuracy of drug resistance prediction (>90%) and adherence by up to 32%, although challenges such as algorithm bias, data privacy, and geographic accessibility remained significant. For visually impaired TB patients, voice technologies such as Google Assistant or Alexa offer non-visual solutions through personalized reminders, text-to-speech pill recognition (85-95% accuracy), and intake confirmation, overcoming the limitations of visual-dependent video-observed therapy (VDOT). Integrating voice AI with digital therapies (DTx) has the potential to improve autonomy and clinical outcomes in resource-limited settings, but requires empirical randomized controlled trials (RCTs) and user training. This study recommends the development of a culturally based hybrid voice therapy-VDOT for Indonesia]]>
Copyrights © 2026
