A multimodal framework for extraction and fusion of satellite images and public health data.

In low- and middle-income countries, the substantial costs associated with traditional data collection pose an obstacle to facilitating decision-making in the field of public health. Satellite imagery offers a potential solution, but the image extraction and analysis can be costly and requires specialized expertise. We introduce SatelliteBench, a scalable framework for satellite image extraction and vector embeddings generation. We also propose a novel multimodal fusion pipeline that utilizes a series of satellite imagery and metadata. The framework was evaluated generating a dataset with a collection of 12,636 images and embeddings accompanied by comprehensive metadata, from 81 municipalities in Colombia between 2016 and 2018. The dataset was then evaluated in 3 tasks: including dengue case prediction, poverty assessment, and access to education. The performance showcases the versatility and practicality of SatelliteBench, offering a reproducible, accessible and open tool to enhance decision-making in public health.

DF-DM: A foundational process model for multimodal data fusion in the artificial intelligence era.

In the big data era, integrating diverse data modalities poses significant challenges, particularly in complex fields like healthcare. This paper introduces a new process model for multimodal Data Fusion for Data Mining, integrating embeddings and the Cross-Industry Standard Process for Data Mining with the existing Data Fusion Information Group model. Our model aims to decrease computational costs, complexity, and bias while improving efficiency and reliability. We also propose "disentangled dense fusion," a novel embedding fusion method designed to optimize mutual information and facilitate dense inter-modality feature interaction, thereby minimizing redundant information. We demonstrate the model's efficacy through three use cases: predicting diabetic retinopathy using retinal images and patient metadata, domestic violence prediction employing satellite imagery, internet, and census data, and identifying clinical and demographic features from radiography images and clinical notes. The model achieved a Macro F1 score of 0.92 in diabetic retinopathy prediction, an R-squared of 0.854 and sMAPE of 24.868 in domestic violence prediction, and a macro AUC of 0.92 and 0.99 for disease prediction and sex classification, respectively, in radiological analysis. These results underscore the Data Fusion for Data Mining model's potential to significantly impact multimodal data processing, promoting its adoption in diverse, resource-constrained settings.

Multidimensional Machine Learning Model to Calculate a COVID-19 Vulnerability Index.

In Colombia, the first case of COVID-19 was confirmed on 6 March 2020. On 13 March 2023, Colombia registered 6,360,780 confirmed positive cases of COVID-19, representing 12.18% of the total population. The National Administrative Department of Statistics (DANE) in Colombia published in 2020 a COVID-19 vulnerability index, which estimates the vulnerability (per city block) of being infected with COVID-19. Unfortunately, DANE did not consider multiple factors that could increase the risk of COVID-19 (in addition to demographic and health), such as environmental and mobility data (found in the related literature). The proposed multidimensional index considers variables of different types (unemployment rate, gross domestic product, citizens' mobility, vaccination data, and climatological and spatial information) in which the incidence of COVID-19 is calculated and compared with the incidence of the COVID-19 vulnerability index provided by DANE. The collection, data preparation, modeling, and evaluation phases of the Cross-Industry Standard Process for Data Mining methodology (CRISP-DM) were considered for constructing the index. The multidimensional index was evaluated using multiple machine learning models to calculate the incidence of COVID-19 cases in the main cities of Colombia. The results showed that the best-performing model to predict the incidence of COVID-19 in Colombia is the Extra Trees Regressor algorithm, obtaining an R-squared of 0.829. This work is the first step toward a multidimensional analysis of COVID-19 risk factors, which has the potential to support decision making in public health programs. The results are also relevant for calculating vulnerability indexes for other viral diseases, such as dengue.

Challenges and solutions for transforming health ecosystems in low- and middle-income countries through artificial intelligence.

Background: Recent studies demonstrate the potential of Artificial Intelligence to support diagnosis, mortality assessment, and clinical decisions in low-and-middle-income countries (LMICs). However, explicit evidence of strategies to overcome the particular challenges for transformed health systems in these countries does not exist. Objective: The present study undertakes a review of research on the current status of artificial intelligence (AI) to identify requirements, gaps, challenges, and possible strategies to strengthen the large, complex, and heterogeneous health systems in LMICs. Design: After introducing the general challenges developing countries face, the methodology of systematic reviews and the meta-analyses extension for scoping reviews (PRISMA-ScR) is introduced according to the preferred reporting items. Scopus and Web of Science databases were used to identify papers published between 2011-2022, from which we selected 151 eligible publications. Moreover, a narrative review was conducted to analyze the evidence in the literature about explicit evidence of strategies to overcome particular AI challenges in LMICs. Results: The analysis of results was divided into two groups: primary studies, which include experimental studies or case studies using or deploying a specific AI solution (n = 129), and secondary studies, including opinion papers, systematic reviews, and papers with strategies or guidelines (n = 22). For both study groups, a descriptive statistical analysis was performed describing their technological contribution, data used, health context, and type of health interventions. For the secondary studies group, an in-deep narrative review was performed, identifying a set of 40 challenges gathered in eight different categories: data quality, context awareness; regulation and legal frameworks; education and change resistance; financial resources; methodology; infrastructure and connectivity; and scalability. A total of 89 recommendations (at least one per challenge) were identified. Conclusion: Research on applying AI and ML to healthcare interventions in LMICs is growing; however, apart from very well-described ML methods and algorithms, there are several challenges to be addressed to scale and mainstream experimental and pilot studies. The main challenges include improving the quality of existing data sources, training and modeling AI solutions based on contextual data; and implementing privacy, security, informed consent, ethical, liability, confidentiality, trust, equity, and accountability policies. Also, robust eHealth environments with trained stakeholders, methodological standards for data creation, research reporting, product certification, sustained investment in data sharing, infrastructures, and connectivity are necessary. Systematic review registration: [https://rb.gy/frn2rz].

Risk Factors for COVID-19: A Systematic Mapping Study.

INTRODUCTION: COVID-19 has affected people in several countries around the world. They experience respiratory symptoms that can be mild, moderate, or severe. Several reviews that characterize the risk factors of COVID-19 have been performed, but most address only risk factors associated with medical conditions, ignoring environmental and sociodemographic-socioeconomic factors. OBJECTIVE: This study aims at characterizing different risk factors in the published literature that influence contagion by COVID-19. METHODS: The review consists of three stages, including a systematic mapping with studies found in the Scopus database, an analysis of results, and finally the identification of relevant COVID-19 risk factors. RESULTS: A map of studies id provided considering two main groups: the type of research and context. Most studies consider risk factors associated with medical conditions, while research on other factors is scarce. CONCLUSIONS: Medical conditions such as diabetes, obesity, cardiovascular disease, hypertension, and factors such as age and sex, appear to be the ones that increase the risk of contracting COVID-19. Further research is needed on environmental, sociodemographic, and socioeconomic risk factors.

Framework for Simultaneous Indoor Localization, Mapping, and Human Activity Recognition in Ambient Assisted Living Scenarios.

Indoor localization and human activity recognition are two important sources of information to provide context-based assistance. This information is relevant in ambient assisted living (AAL) scenarios, where older adults usually need supervision and assistance in their daily activities. However, indoor localization and human activity recognition have been mostly considered isolated problems. This work presents and evaluates a framework that takes advantage of the relationship between location and activity to simultaneously perform indoor localization, mapping, and human activity recognition. The proposed framework provides a non-intrusive configuration, which fuses data from an inertial measurement unit (IMU) placed in the person's shoe, with proximity and human activity-related data from Bluetooth low energy beacons (BLE) deployed in the indoor environment. A variant of the simultaneous location and mapping (SLAM) framework was used to fuse the location and human activity recognition (HAR) data. HAR was performed using data streaming algorithms. The framework was evaluated in a pilot study, using data from 22 people, 11 young people, and 11 older adults (people aged 65 years or older). As a result, seven activities of daily living were recognized with an F1 score of 88%, and the in-door location error was 0.98 ± 0.36 m for the young and 1.02 ± 0.24 m for the older adults. Furthermore, there were no significant differences between the groups, indicating that our proposed method works adequately in broad age ranges.

Transformation of Health and Social Care Systems-An Interdisciplinary Approach Toward a Foundational Architecture.

Objective: For realizing pervasive and ubiquitous health and social care services in a safe and high quality as well as efficient and effective way, health and social care systems have to meet new organizational, methodological, and technological paradigms. The resulting ecosystems are highly complex, highly distributed, and highly dynamic, following inter-organizational and even international approaches. Even though based on international, but domain-specific models and standards, achieving interoperability between such systems integrating multiple domains managed by multiple disciplines and their individually skilled actors is cumbersome. Methods: Using the abstract presentation of any system by the universal type theory as well as universal logics and combining the resulting Barendregt Cube with parameters and the engineering approach of cognitive theories, systems theory, and good modeling best practices, this study argues for a generic reference architecture model moderating between the different perspectives and disciplines involved provide on that system. To represent architectural elements consistently, an aligned system of ontologies is used. Results: The system-oriented, architecture-centric, and ontology-based generic reference model allows for re-engineering the existing and emerging knowledge representations, models, and standards, also considering the real-world business processes and the related development process of supporting IT systems for the sake of comprehensive systems integration and interoperability. The solution enables the analysis, design, and implementation of dynamic, interoperable multi-domain systems without requesting continuous revision of existing specifications.

Correction: Evaluating the impact of multivariate imputation by MICE in feature selection.

[This corrects the article DOI: 10.1371/journal.pone.0254720.].

Evaluating the impact of multivariate imputation by MICE in feature selection.

Handling missing values is a crucial step in preprocessing data in Machine Learning. Most available algorithms for analyzing datasets in the feature selection process and classification or estimation process analyze complete datasets. Consequently, in many cases, the strategy for dealing with missing values is to use only instances with full data or to replace missing values with a mean, mode, median, or a constant value. Usually, discarding missing samples or replacing missing values by means of fundamental techniques causes bias in subsequent analyzes on datasets. AIM: Demonstrate the positive impact of multivariate imputation in the feature selection process on datasets with missing values. RESULTS: We compared the effects of the feature selection process using complete datasets, incomplete datasets with missingness rates between 5 and 50%, and imputed datasets by basic techniques and multivariate imputation. The feature selection algorithms used are well-known methods. The results showed that the datasets imputed by multivariate imputation obtained the best results in feature selection compared to datasets imputed by basic techniques or non-imputed incomplete datasets. CONCLUSIONS: Considering the results obtained in the evaluation, applying multivariate imputation by MICE reduces bias in the feature selection process.

Recalibration of deep learning models for abnormality detection in smartphone-captured chest radiograph.

Image-based teleconsultation using smartphones has become increasingly popular. In parallel, deep learning algorithms have been developed to detect radiological findings in chest X-rays (CXRs). However, the feasibility of using smartphones to automate this process has yet to be evaluated. This study developed a recalibration method to build deep learning models to detect radiological findings on CXR photographs. Two publicly available databases (MIMIC-CXR and CheXpert) were used to build the models, and four derivative datasets containing 6453 CXR photographs were collected to evaluate model performance. After recalibration, the model achieved areas under the receiver operating characteristic curve of 0.80 (95% confidence interval: 0.78-0.82), 0.88 (0.86-0.90), 0.81 (0.79-0.84), 0.79 (0.77-0.81), 0.84 (0.80-0.88), and 0.90 (0.88-0.92), respectively, for detecting cardiomegaly, edema, consolidation, atelectasis, pneumothorax, and pleural effusion. The recalibration strategy, respectively, recovered 84.9%, 83.5%, 53.2%, 57.8%, 69.9%, and 83.0% of performance losses of the uncalibrated model. We conclude that the recalibration method can transfer models from digital CXRs to CXR photographs, which is expected to help physicians' clinical works.

Multi-Dimensional Dataset of Open Data and Satellite Images for Characterization of Food Security and Nutrition.

BACKGROUND: Nutrition is one of the main factors affecting the development and quality of life of a person. From a public health perspective, food security is an essential social determinant for promoting healthy nutrition. Food security embraces four dimensions: physical availability of food, economic and physical access to food, food utilization, and the sustainability of the dimensions above. Integrally addressing the four dimensions is vital. Surprisingly most of the works focused on a single dimension of food security: the physical availability of food. OBJECTIVE: The paper proposes a multi-dimensional dataset of open data and satellite images to characterize food security in the department of Cauca, Colombia. METHODS: The food security dataset integrates multiple open data sources; therefore, the Cross-Industry Standard Process for Data Mining methodology was used to guide the construction of the dataset. It includes sources such as population and agricultural census, nutrition surveys, and satellite images. RESULTS: An open multidimensional dataset for the Department of Cauca with 926 attributes and 9 rows (each row representing a Municipality) from multiple sources in Colombia, is configured. Then, machine learning models were used to characterize food security and nutrition in the Cauca Department. As a result, The Food security index calculated for Cauca using a linear regression model (Mean Absolute Error of 0.391) is 57.444 in a range between 0 and 100, with 100 the best score. Also, an approach for extracting four features (Agriculture, Habitation, Road, Water) of satellite images were tested with the ResNet50 model trained from scratch, having the best performance with a macro-accuracy, macro-precision, macro-recall, and macro-F1-score of 91.7, 86.2, 66.91, and 74.92%, respectively. CONCLUSION: It shows how the CRISP-DM methodology can be used to create an open public health data repository. Furthermore, this methodology could be generalized to other types of problems requiring the creation of a dataset. In addition, the use of satellite images presents an alternative for places where data collection is challenging. The model and methodology proposed based on open data become a low-cost and effective solution that could be used by decision-makers, especially in developing countries, to support food security planning.

Simultaneous Indoor Pedestrian Localization and House Mapping Based on Inertial Measurement Unit and Bluetooth Low-Energy Beacon Data.

Indoor location estimation is crucial to provide context-based assistance in home environments. In this study, a method for simultaneous indoor pedestrian localization and house mapping is proposed and evaluated. The method fuses a person's movement data from an Inertial Measurement Unit (IMU) with proximity and activity-related data from Bluetooth Low-Energy (BLE) beacons deployed in the indoor environment. The person's and beacons' localization is performed simultaneously using a combination of particle and Kalman Filters. We evaluated the method using data from eight participants who performed different activities in an indoor environment. As a result, the average participant's localization error was 1.05 ± 0.44 m, and the average beacons' localization error was 0.82 ± 0.24 m. The proposed method is able to construct a map of the indoor environment by localizing the BLE beacons and simultaneously locating the person. The results obtained demonstrate that the proposed method could point to a promising roadmap towards the development of simultaneous localization and home mapping system based only on one IMU and a few BLE beacons. To the best of our knowledge, this is the first method that includes the beacons' data movement as activity-related events in a method for pedestrian Simultaneous Localization and Mapping (SLAM).

Personalized Tracking of Physical Activity in Children Using a Wearable Heart Rate Monitor.

Serious games are video games that are intended to support learning while entertaining. They are considered valuable tools to improve user-specific skills or facilitate educational or therapeutic processes, especially in children. One of the disadvantages of computer games, in general, is their promotion of sedentary habits, considered as a significant risk factor for developing diseases such as obesity and hypertension. Exergames are serious games created to overcome the disadvantages of traditional computer games by promoting physical activity while playing. This study describes the development and evaluation of an adaptive component to monitor physical activity in children while using an exergame. The system is based on wearable technology to measure heart rate and perform real-time customizations in the exergame. To evaluate the adaptive component, an experiment was conducted with 30 children between 5 and 7 years of age, where the adaptive system was contrasted with a conventional interactive system (an exergame without adaptive component). It was demonstrated that the computer game, using the adaptive component, was able to change in real-time some of its functionalities based on the user characteristics. Increased levels of heart rate and caloric expenditure were significant in some of the game scenarios using the adaptive component. Although a formal user experience evaluation was not performed, excellent game playability and adherence by users were observed.

Intelligent Telehealth System To Support Epilepsy Diagnosis.

BACKGROUND: Availability and opportunity of epilepsy diagnostic services is a significant challenge, especially in developing countries with a low number of neurologists. The most commonly used test to diagnose epilepsy is electroencephalogram (EEG). A typical EEG recording lasts for 20 to 30 minutes; however, a specialist requires much more time to read it. Furthermore, no evidence was found in the literature on open-source systems for the cost-effective management of patient information using electronic health records (EHR) that adequately integrate EEG analysis for automatic identification of abnormal signals. OBJECTIVE: To develop an integrated open-source EHR system for the management of the patients' personal, clinical, and EEG data, and for automatic identification of abnormal EEG signals. METHODS: The core of the system is an EHR and telehealth service based on the OpenMRS platform. On top of that, we developed an intelligent component to automatically detect abnormal segments of EEG tests using machine learning algorithms, as well as a service to annotate and visualize abnormal segments in EEG signals. Finally, we evaluated the intelligent component and the integrated system using precision, recall, and accuracy metrics. RESULTS: The system allowed to manage patients' information properly, store and manage the EEG tests recorded with a medical EEG device, and to detect abnormal segments of signals with a precision of 85.10%, a recall of 97.16%, and an accuracy of 99.92%. CONCLUSION: Digital health is a multidisciplinary field of research in which artificial intelligence is playing a significant role in boosting traditional health services. Notably, the developed system could significantly reduce the time a neurologist spends in the reading of an EEG for the diagnosis of epilepsy, saving approximately 65-75% of the time consumed. It can be used in a telehealth environment. In this way, the availability and provision of diagnostic services for epilepsy management could be improved, especially in developing countries where the number of neurologists is low.

Ontology-Based and Architecture-Based Method for the Development of Interoperable Care Systems for Type 2 Diabetes Mellitus.

Processes like the care of type 2 diabetes mellitus patients require support by information systems considering the heterogeneity of the actors from different domains involved, enabling harmonization and integration of their specific methodologies and knowledge representation approaches towards interdisciplinary cooperation. Currently, the development of systems starts from the simplified information world, ignoring the aforementioned heterogeneity and specificity of real-world processes. This paper aims to demonstrate the feasibility of developing an adaptive, interoperable and intelligent system that supports the major aspects of type 2 diabetes mellitus care based on the Generic Component Model as formal methodology for modelling universal systems. The result is a deployable solution based on a formal representation of the diabetes care system, its objectives, and the intended business process. The implemented system enables reasoning over the data, inferring medical diagnosis. The effectiveness of the inference was evaluated, obtaining an F-measure of 0.89. The methods presented in this paper helps to build high quality models based on computation-independent aspects, which enable the construction of knowledge-based adaptive, intelligent and interoperable eHealth systems.

Epileptic spikes detector in pediatric EEG based on matched filters and neural networks.

The electroencephalogram (EEG) is a tool for diagnosing epilepsy; by analyzing it, neurologists can identify alterations in brain activity associated with epilepsy. However, this task is not always easy to perform because of the duration of the EEG or the subjectivity of the specialist in detecting alterations. AIM: To propose the use of an epileptic spike detector based on a matched filter and a neural network for supporting the diagnosis of epilepsy through a tool capable of automatically detecting spikes in pediatric EEGs. RESULTS: Automatic detection of spikes from an EEG waveform involved the creation of an epileptic spike template. The template was used in order to detect spikes by using a matched filter, and each spike detected was confirmed by a Neural Network to improve sensitivity and specificity. Thus, the detector developed achieved a sensitivity of 99.96% which is better than the range of what has been reported in the literature (82.68% and 94.4%), and a specificity of 99.26%, improving the specificity found in the best-reviewed studies. CONCLUSIONS: Considering the results obtained in the evaluation, the solution becomes a promising alternative to support the automatic identification of epileptic spikes by neurologists.

Indoor Trajectory Reconstruction of Walking, Jogging, and Running Activities Based on a Foot-Mounted Inertial Pedestrian Dead-Reckoning System.

The evaluation of trajectory reconstruction of the human body obtained by foot-mounted Inertial Pedestrian Dead-Reckoning (IPDR) methods has usually been carried out in controlled environments, with very few participants and limited to walking. In this study, a pipeline for trajectory reconstruction using a foot-mounted IPDR system is proposed and evaluated in two large datasets containing activities that involve walking, jogging, and running, as well as movements such as side and backward strides, sitting, and standing. First, stride segmentation is addressed using a multi-subsequence Dynamic Time Warping method. Then, detection of Toe-Off and Mid-Stance is performed by using two new algorithms. Finally, stride length and orientation estimation are performed using a Zero Velocity Update algorithm empowered by a complementary Kalman filter. As a result, the Toe-Off detection algorithm reached an F-score between 90% and 100% for activities that do not involve stopping, and between 71% and 78% otherwise. Resulting return position errors were in the range of 0.5% to 8.8% for non-stopping activities and 8.8% to 27.4% otherwise. The proposed pipeline is able to reconstruct indoor trajectories of people performing activities that involve walking, jogging, running, side and backward walking, sitting, and standing.

Analysis of Relevant Features from Photoplethysmographic Signals for Atrial Fibrillation Classification.

Atrial Fibrillation (AF) is the most common cardiac arrhythmia found in clinical practice. It affects an estimated 33.5 million people, representing approximately 0.5% of the world's population. Electrocardiogram (ECG) is the main diagnostic criterion for AF. Recently, photoplethysmography (PPG) has emerged as a simple and portable alternative for AF detection. However, it is not completely clear which are the most important features of the PPG signal to perform this process. The objective of this paper is to determine which are the most relevant features for PPG signal analysis in the detection of AF. This study is divided into two stages: (a) a systematic review carried out following the Preferred Reporting Items for a Systematic Review and Meta-analysis of Diagnostic Test Accuracy Studies (PRISMA-DTA) statement in six databases, in order to identify the features of the PPG signal reported in the literature for the detection of AF, and (b) an experimental evaluation of them, using machine learning, in order to determine which have the greatest influence on the process of detecting AF. Forty-four features were found when analyzing the signal in the time, frequency, or time-frequency domains. From those 44 features, 27 were implemented, and through machine learning, it was found that only 11 are relevant in the detection process. An algorithm was developed for the detection of AF based on these 11 features, which obtained an optimal performance in terms of sensitivity (98.43%), specificity (99.52%), and accuracy (98.97%).

Adaptation Component Based on Wearable Technology to Support Personalized Tracking of Physical Activity in Children.

Obesity is a chronic disease characterized by the accumulation of body fat. School-age children obesity is one of the most serious public health challenges since it progressively becomes a risk factor in adulthood. Recent studies and technological innovations have demonstrated the feasibility of game-based interventions for promoting physical activity among children. Nevertheless, there is a gap in fitting the system specifications to specific user profiles. This paper aims to present the development of an adaptation component for a particular exergame based on wearable technology that measures heart rate to support a personalized tracking system of the physical activity. The employed methodology was the General Adaptivity Model (GAM) that provides guidelines for the designing process of adaptivity models, incorporating user modeling and personalization in existing or new interactive systems. For validation purposes, an experiment was conducted at a primary school with thirty subjects aged between five to seven years to test the effectiveness of a user adaptive system against a conventional interactive system for the promotion of physical activity. Results indicate that the developed system was able to change its behavior according to the variations of the heart rate and therefore encouraging users to perform higher/lower physical activity levels.