Technologies for Monitoring Lifestyle Habits Related to Brain Health: A Systematic Review

BrainSegFounder: Towards 3D foundation models for neuroimage segmentation

The burgeoning field of brain health research increasingly leverages artificial intelligence (AI) to analyze and interpret neuroimaging data. Medical foundation models have shown promise of superior performance with better sample efficiency. This work introduces a novel approach towards creating 3-dimensional (3D) medical foundation models for multimodal neuroimage segmentation through self-supervised training. Our approach involves a novel two-stage pretraining approach using vision transformers. The first stage encodes anatomical structures in generally healthy brains from the large-scale unlabeled neuroimage dataset of multimodal brain magnetic resonance imaging (MRI) images from 41,400 participants. This stage of pertaining focuses on identifying key features such as shapes and sizes of different brain structures. The second pretraining stage identifies disease-specific attributes, such as geometric shapes of tumors and lesions and spatial placements within the brain. This dual-phase methodology significantly reduces the extensive data requirements usually necessary for AI model training in neuroimage segmentation with the flexibility to adapt to various imaging modalities. We rigorously evaluate our model, BrainSegFounder, using the Brain Tumor Segmentation (BraTS) challenge and Anatomical Tracings of Lesions After Stroke v2.0 (ATLAS v2.0) datasets. BrainSegFounder demonstrates a significant performance gain, surpassing the achievements of the previous winning solutions using fully supervised learning. Our findings underscore the impact of scaling up both the model complexity and the volume of unlabeled training data derived from generally healthy brains. Both of these factors enhance the accuracy and predictive capabilities of the model in neuroimage segmentation tasks. Our pretrained models and code are at https://github.com/lab-smile/BrainSegFounder.

Toward Telemonitoring in Immune-Mediated Inflammatory Diseases: Protocol for a Mixed Attention Model Study

BACKGROUND

Rheumatic and musculoskeletal diseases (RMDs) are chronic diseases that may alternate between asymptomatic periods and flares. These conditions require complex treatments and close monitoring by rheumatologists to mitigate their effects and improve the patient's quality of life. Often, delays in outpatient consultations or the patient's difficulties in keeping appointments make such close follow-up challenging. For this reason, it is very important to have open communication between patients and health professionals. In this context, implementing telemonitoring in the field of rheumatology has great potential, as it can facilitate the close monitoring of patients with RMDs. The use of these tools helps patients self-manage certain aspects of their disease. This could result in fewer visits to emergency departments and consultations, as well as enable better therapeutic compliance and identification of issues that would otherwise go unnoticed.

OBJECTIVE

The main objective of this study is to evaluate the implementation of a hybrid care model called the mixed attention model (MAM) in clinical practice and determine whether its implementation improves clinical outcomes compared to conventional follow-up.

METHODS

This is a multicenter prospective observational study involving 360 patients with rheumatoid arthritis (RA) and spondylarthritis (SpA) from 5 Spanish hospitals. The patients will be followed up by the MAM protocol, which is a care model that incorporates a digital tool consisting of a mobile app that patients can use at home and professionals can review asynchronously to detect incidents and follow patients' clinical evolution between face-to-face visits. Another group of patients, whose follow-up will be conducted in accordance with a traditional face-to-face care model, will be assessed as the control group. Sociodemographic characteristics, treatments, laboratory parameters, assessment of tender and swollen joints, visual analog scale for pain, and electronic patient-reported outcome (ePRO) reports will be collected for all participants. In the MAM group, these items will be self-assessed via both the mobile app and during face-to-face visits with the rheumatologist, who will do the same for patients included in the traditional care model. The patients will be able to report any incidence related to their disease or treatment through the mobile app.

RESULTS

Participant recruitment began in March 2024 and will continue until December 2024. The follow-up period will be extended by 12 months for all patients. Data collection and analysis are scheduled for completion in December 2025.

CONCLUSIONS

This paper aims to provide a detailed description of the development and implementation of a digital solution, specifically an MAM. The goal is to achieve significant economic and psychosocial impact within our health care system by enhancing control over RMDs.

TRIAL REGISTRATION

ClinicalTrials.gov NCT06273306; https://clinicaltrials.gov/ct2/show/NCT06273306.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

PRR1-10.2196/55829.

Intelligent Coaching Assistant for the Promotion of Healthy Habits in a Multidomain mHealth-Based Intervention for Brain Health

Brain Health is defined as the development and preservation of optimal brain integrity and neural network functioning for a given age. Recent studies have related healthy habits with better maintenance of brain health across the lifespan. As a part of the Barcelona Brain Health Initiative (BBHI), a mHealth platform has been developed with the purpose of helping people to improve and monitor their healthy habits, facilitating the delivery of health coaching strategies. A decision support system (DSS), named Intelligent Coaching Assistant (ICA), has been developed to ease the work of professional brain health coaches, helping them design and monitor adherence to multidomain interventions in a more efficient manner. Personalized recommendations are based on users’ current healthy habits, individual preferences, and motivational aspects. Taking these inputs, an initial user profile is defined, and the ICA applies an algorithm for determining the most suitable personalized intervention plan. An initial validation has been done focusing on assessing the feasibility and usability of the solution, involving 20 participants for three weeks. We conclude that this kind of technology-based intervention is feasible and implementable in real-world settings. Importantly, the personalized intervention proposal generated by the DSS is feasible and its acceptability and usability are high.

Special Issue "Body Sensors Networks for E-Health Applications"

Body Sensor Networks (BSN) have emerged as a particularization of Wireless Sensor Networks (WSN) in the context of body monitoring environments, closely linked to healthcare applications. These networks are made up of smart biomedical sensors that allow the monitoring of physiological parameters and serve as the basis for e-Health applications. This Special Issue collects some of the latest developments in the field of BSN related to new developments in biomedical sensor technologies, the design and experimental characterization of on-body/in-body antennas and new communication protocols for BSN, including some review studies.