17 October 2017 in PLOS ONE
This paper presents the concept of a social autonomous agent to conceptualize such Autonomous Vehicles (AVs), which interacts with other AVs using social manners similar to human behavior. The presented AVs also have the capability of predicting intentions, i.e. mentalizing and copying the actions of each other, i.e. mirroring. Exploratory Agent Based Modeling (EABM) level of the Cognitive Agent Based Computing (CABC) framework has been utilized to design the proposed social agent. Furthermore, to emulate the functionality of mentalizing and mirroring modules of proposed social agent, a tailored mathematical model of the Richardson’s arms race model has also been presented. The performance of the proposed social agent has been validated at two levels–firstly it has been simulated using NetLogo, a standard agent-based modeling tool and also, at a practical level using a prototype AV. The simulation results have confirmed that the proposed social agent-based collision avoidance strategy is 78.52% more efficient than Random walk based collision avoidance strategy in congested flock-like topologies. Whereas practical results have confirmed that the proposed scheme can avoid rear end and lateral collisions with the efficiency of 99.876% as compared with the IEEE 802.11n-based existing state of the art mirroring neuron-based collision avoidance scheme.
22 December 2021 in Journal of Affective Disorders
Background: Major depressive disorder (MDD) is predominantly managed in primary care. However, primary care providers (PCPs) may not consistently follow evidence-based treatment algorithms, leading to variable patient management that can impact outcomes. Methods: We retrospectively analyzed adult patients with MDD seen at Geisinger, an integrated health system. Utilizing electronic health record (EHR) data, we classified patients as having MDD based on International Classification of Disease (ICD)-9/10 codes or a Patient Health Questionnaire (PHQ)-9 score ≥5. Outcomes assessed included time to first visit with a PCP or behavioral health specialist following diagnosis, antidepressant medication switching, persistence, healthcare resource utilization (HRU), and treatment costs. Results: Among the 38,321 patients with MDD managed in primary care in this study, significant delays between diagnosis with antidepressant prescribing and follow-up PCP visits were observed. There was also considerable variation in care following diagnosis. Overall, 34.9% of patients with an ICD-9/10 diagnosis of MDD and 41.3% with a PHQ-9 score ≥15 switched antidepressants. An ICD-9/10 diagnosis, but not moderately severe to severe depression, was associated with higher costs and HRU. More than 75% of patients with MDD discontinued antidepressant medication within 6 months. Limitations: The study population was comparable with other real-world studies of MDD, but study limitations include its retrospective nature and reliance on the accuracy of EHRs. Conclusions: Management of patients with MDD in a primary care setting is variable. Addressing these gaps will have important implications for ensuring optimal patient management, which may reduce HRU and treatment medication costs, and improve treatment persistence.
1 January 2022 in Computers, Materials & Continua
Deep learning techniques, particularly convolutional neural networks (CNNs), have exhibited remarkable performance in solving vision-related problems, especially in unpredictable, dynamic, and challenging environments. In autonomous vehicles, imitation-learning-based steering angle prediction is viable due to the visual imagery comprehension of CNNs. In this regard, globally, researchers are currently focusing on the architectural design and optimization of the hyperparameters of CNNs to achieve the best results. Literature has proven the superiority of metaheuristic algorithms over the manual-tuning of CNNs. However, to the best of our knowledge, these techniques are yet to be applied to address the problem of imitation-learning-based steering angle prediction. Thus, in this study, we examine the application of the bat algorithm and particle swarm optimization algorithm for the optimization of the CNN model and its hyperparameters, which are employed to solve the steering angle prediction problem. To validate the performance of each hyperparameters’ set and architectural parameters’ set, we utilized the Udacity steering angle dataset and obtained the best results at the following hyperparameter set: optimizer, Adagrad; learning rate, 0.0052; and nonlinear activation function, exponential linear unit. As per our findings, we determined that the deep learning models show better results but require more training epochs and time as compared to shallower ones. Results show the superiority of our approach in optimizing CNNs through metaheuristic algorithms as compared with the manual-tuning approach. Infield testing was also performed using the model trained with the optimal architecture, which we developed using our approach.
1 July 2024 in Journal of Medical Economics
Aims: Patients with inborn errors of immunity (IEI) are predisposed to severe recurrent/chronic infections, and often require hospitalization, resulting in substantial burden to patients/healthcare systems. While immunoglobulin replacement therapies (IgRTs) are the standard first-line treatment for most forms of IEI, limited real-world data exist regarding clinical characteristics and treatment costs for patients with IEI initiating such treatment. This retrospective analysis examined infection and treatment characteristics in US patients with IEI initiating IgRT with immune globulin infusion (human), 10% (IG10%). Healthcare resource utilization (HCRU) and associated costs before and after treatment initiation were compared. Additionally, the impact of COVID-19 on infection diagnoses was evaluated.Methods: Patients with IEI initiating IG10% between July 2012-August 2019 were selected from Merative® MarketScan® Databases using diagnosis/prescription codes. Patients were followed 6 months before and after first IG10% claim date. Demographic and clinical characteristics were described. Treatment characteristics and HCRU before and after IG10% initiation were compared. Infection diagnoses during 2020 and 2019 (March-December) were compared.Results: The study included 1,497 patients with IEI diagnoses (mean age = 43.4 years) initiating IG10%, with frequently reported comorbidities like asthma (32.1%). Following IG10% initiation, fewer severe infection diagnoses (11.6% vs 19.9%), fewer infection-related inpatient (10.8% vs 19.5%) and outpatient services (71.6% vs 79.9%), and lower infection-related total healthcare costs ($7,849 vs $13,995; P < 0.001)-driven by lower inpatient costs ($2,746 vs $9,900)-were observed than before. Fewer patients had infection diagnoses during COVID-19 (22.8%) than the prior year (31.2%).Conclusion: Patients with IEI are susceptible to severe infections leading to high disease burden and treatment costs. Following IG10% initiation, we observed fewer infections, lower infection-related treatment costs, and shift in care (inpatient to outpatient), leading to significant cost savings. Among patients with IEI, 27% fewer infection diagnoses were observed during the early COVID-19 lockdown period than the prior year.
19 March 2026 in BRAIN. BROAD RESEARCH IN ARTIFICIAL INTELLIGENCE AND NEUROSCIENCE
Predicting brake light activation in semi-autonomous vehicles (SAVs) is critical for improving road safety and optimising adaptive cruise control (ACC) systems. Despite the remarkable performance of deep learning-based object detectors such as YOLOv8, their inherent opacity in decision-making processes limits transparency, interpretability, and user trust factors essential for the deployment of safety-critical systems. This study introduces a novel, explainable framework that integrates state-of-the-art attribution-based explainability techniques, including EigenCAM, EigenGrad-CAM, LayerCAM, and HiResCAM, into the YOLOv8 architecture. The proposed framework systematically analyses activation patterns within the model, generating fine-grained saliency maps that highlight the spatial regions most influential in brake light detection. Using a real-world vehicle dataset comprising diverse lighting and environmental conditions, the model is fine-tuned to predict binary brake light states (car_BrakeOn and car_BrakeOff). Comparative experiments demonstrate the ability of these CAM-based methods to provide interpretable visual explanations while maintaining detection accuracy. This study is the first to explore the integration of explainability techniques within YOLO-based brake light detection systems for semi-autonomous vehicles (SAE Levels 2 and 3), addressing a critical gap in the literature. By bridging the divide between black-box AI models and human-understandable reasoning, this research promotes transparency, accountability, and user trust in AI-driven perception systems for intelligent transportation. The findings contribute a foundational step toward the broader adoption of explainable and reliable AI in safety-critical SAV applications.