Background/Objectives: Image-based food energy estimation is essential for user-friendly food tracking applications, enabling individuals to monitor their dietary intake through smartphones or AR devices. However, existing deep learning approaches struggle to recognize a wide variety of food items, due to the labor-intensive nature of data annotation. Multimodal Large Language Models (MLLMs) possess extensive knowledge and human-like reasoning abilities, making them a promising approach for image-based food energy estimation. Nevertheless, their ability to accurately estimate food energy is hindered by limitations in recognizing food size, a critical factor in energy content assessment. Methods: To address this challenge, we propose two approaches: fine-tuning, and volume-aware reasoning with fine-grained estimation prompting. Results: Experimental results on the Nutrition5k dataset demonstrated the effectiveness of these approaches in improving estimation accuracy. We also validated the effectiveness of adapting LoRA to enhance food energy estimation performance. Conclusions: These findings highlight the potential of MLLMs for image-based dietary assessment and emphasize the importance of integrating volume-awareness into food energy estimation models.

Background/Objectives: Although information and communication technology (ICT) offers opportunities to address challenges, evidence among frail populations is limited. We aimed to evaluate the effectiveness and feasibility of an ICT-based intervention incorporating an artificial intelligence (AI)-assisted smartphone dietary application and group communication tools to improve dietary quality and social connection among community-dwelling older adults with frailty. Methods: A non-randomized, quasi-experimental study was conducted among 29 older adults (≥65 years) in Tokyo, Japan. Participants were assigned to the intervention (n = 11) or control (n = 18) group. The 3-month intervention included weekly photo uploads of meals via an AI-based dietary application providing automated image analysis and personalized feedback, supervised by registered dietitians, along with peer communication through a group chat. The primary outcome was dietary quality. The secondary outcomes included body weight, body mass index (BMI), skin carotenoid score, and loneliness. Results: The adjusted Japanese Food Guide Spinning Top Score at 3-month follow-up was 49.0 (standard error [SE] = 2.6) and 39.5 (SE = 2.0) in the intervention and control groups, respectively. The adjusted mean difference between groups was +9.5 (95% confidence interval: 2.3 to 16.7, p = 0.01). After using analysis of covariance for adjusting for respective baseline values, age, education status, and antihypertension drug use, no statistically significant between-group differences were observed at 3-month follow-up for any secondary outcomes. Conclusions: AI-based dietary intervention and peer communication effectively improved dietary quality among older adults, highlighting the potential of such an intervention to promote healthier eating habits in this population.