Explore the vast range of titles available.

Non-Communicable Diseases (NCDs) are now a leading cause of mortality and morbidity globally, and mental illness is a significant part of it. In India, the treatment gap for common mental disorders is over 80%. In order to bridge this gap, mental health treatment models recommend task-shifting to non-specialists and integration of mental health care into general healthcare services. Other NCDs are being managed effectively by non-specialist healthcare workers (HCWs) at primary care, and mental illness and substance misuse are highly comorbid with other NCDs; hence, integrating mental health care within the NCD services and care framework seems logically feasible and effective. However, country-specific characteristics pose a significant challenge to the implementation of integrated care for mental disorders and NCDs. The primary objective of this study includes the development and implementation of a service delivery model that would result in at least 70% coverage of screening, linkage to care, and management of common mental disorders and substance use disorders (MSUD) among persons seeking care for NCDs at public health facilities. Secondary objectives include assessment of the feasibility of adoption of the implementation model by the health care system and to evaluate the cost of the mental health service strengthening intervention package from the health system’s and the patient’s perspectives. It will be a multi-site implementation research study, employing a mixed-methods quasi-experimental, within-site, three-phase, single-arm, interrupted time series design. The implementation model comprises screening, treatment, and linkage of mental health services integrated into NCD care in at least three blocks in each of the seven selected districts of the seven selected states of India, which are geographically far apart. The expected outcome would be to increase the proportion of patients screened and managed for MSUDs among persons seeking care for NCDs at the public health facilities. The results of this implementation research will provide a roadmap for scaling up of integrated MSUDs services within general healthcare.

Background and objectives There is an increased burden of mental disorders, including substance use disorders (MSUD) in low- and middle-income countries (LMICs). Digital technology offers a viable avenue to scale up the services for MSUD in such settings. This review aims to synthesize existing evidence from systematic reviews and meta-analyses on the use of mobile applications for mental disorders, including substance use disorders in LMICs. Methods A comprehensive search for review articles and meta-analyses was conducted in January 2025 and encompassed all relevant studies published until then using the electronic databases PubMed, Scopus, and Embase. Rayyan software was employed to remove duplicates. Data extraction included information such as publication date, title, author, country, number of participants, study design, data collection procedures, instruments/tools used, the profile of healthcare workers, the profile of patients, name of the mobile application, use of the mobile application and privacy policy. Evaluation of risk of bias and quality assessment for the included studies was carried out using the Cochrane tool for assessing risk of bias RoB 2, ROBINS-I for nonrandomized studies, and the Mixed Methods Appraisal Tool. The quality of the included systematic reviews was assessed using A MeaSurement Tool to Assess systematic Reviews. Results Twenty-three individual studies that met the eligibility criteria were included in the review. The effectiveness of mobile applications for mental health in LMICs varied significantly across studies. Despite the high burden of substance use disorders in LMICs, only a few studies evaluated mobile applications targeting substance use disorders. While mobile applications offer numerous advantages, significant barriers to widespread implementation remain. Conclusions The findings of this review have significant implications for future research, policy, and practice. Given the potential of mobile applications to improve mental health care in LMICs, efforts should focus on enhancing their cultural relevance, usability, and long-term effectiveness.

IntroductionDespite implementation of the National Programme for Prevention and Control of Non-Communicable Diseases (NP-NCD), screening coverage for oral, breast and cervical cancers remains below 2%. Screening quality is inadequately addressed and delays in diagnosis and treatment initiation continue to persist. This multisite implementation research aims to improve district-level coverage and quality of screening, early diagnosis and timeliness of treatment initiation through a model co-developed within the NP-NCD context.Methods and analysisThe study will be conducted in three phases across seven districts in diverse regions of India. In phase I (formative), the current status, barriers and facilitators of cancer screening, diagnosis and treatment initiation under NP-NCD will be assessed. In phase II (optimisation), a model (package of implementation strategies) will be co-developed and iteratively optimised with multistakeholder engagement at the subdistrict level to improve screening coverage and quality and strengthen the referral system for early diagnosis and treatment initiation. In phase III (scale-up and evaluation), the model will be implemented at the district level and evaluated for improvements in screening, early diagnosis and treatment initiation. A convergent mixed-methods design will be used, incorporating household surveys, facility assessments and stakeholder interviews. Implementation Research Logic Model will guide planning, execution and evaluation in the present study. Determinants of screening coverage and quality, early diagnosis and treatment initiation will be assessed using the Consolidated Framework for Implementation Research. Implementation strategies for the model will be finalised using the Expert Recommendations for Implementing Change framework. Implementation and service outcomes will be evaluated using the Reach, Effectiveness, Adoption, Implementation and Maintenance framework.Ethics and disseminationEthical approval has been obtained from all study sites. The study findings will be disseminated at the state, national and global levels through meetings and conferences and submitted to a peer-reviewed journal for publication.Trial registration numberCTRI/2025/08/092672.

Polycystic Ovary Syndrome (PCOS) is one of the most widespread endocrine and metabolic disorders affecting women of reproductive age. Major symptoms include hyperandrogenism, polycystic ovary, irregular menstruation cycle, excessive hair growth, etc., which sometimes may lead to more severe complications like infertility, pregnancy complications and other co-morbidities such as diabetes, hypertension, sleep apnea, etc. Early detection and effective management of PCOS are essential to enhance patients' quality of life and reduce the chances of associated health complications. Artificial intelligence (AI) techniques have recently emerged as a popular methodology in the healthcare industry for diagnosing and managing complex diseases such as PCOS. AI utilizes machine learning algorithms to analyze ultrasound images and anthropometric and biochemical test result data to diagnose PCOS quickly and accurately. AI can assist in integrating different data sources, such as patient histories, lab findings, and medical records, to present a clear and complete picture of an individual's health. This information can help the physician make more informed and efficient diagnostic decisions. This review article provides a comprehensive analysis of the evolving role of AI in various aspects of the management of PCOS, with a major focus on AI-based diagnosis tools. Graphical Abstract

Recent advances in Artificial Intelligence (AI) have brought AI closer to laypeople than ever before. This leads to a pervasive problem: how would a user ascertain whether an AI system will be safe, reliable, or useful in a given situation? This problem becomes particularly challenging when it is considered that most autonomous systems are not designed by their users; the internal software of these systems may be unavailable or difficult to understand; and the functionality of these systems may even change from initial specifications as a result of learning. To overcome these challenges, this dissertation proposes a paradigm for third-party autonomous assessment of black-box taskable AI systems. The four main desiderata of such assessment systems are: (i) interpretability: generating a description of the AI system's functionality in a language that the target user can understand; (ii) correctness: ensuring that the description of AI system's working is accurate; (iii) generalizability creating a solution approach that works well for different types of AI systems; and (iv) minimal requirements: creating an assessment system that does not place complex requirements on AI systems to support the third-party assessment, otherwise the manufacturers of AI system's might not support such an assessment. To satisfy these properties, this dissertation presents algorithms and requirements that would enable user-aligned autonomous assessment that helps the user understand the limits of a black-box AI system's safe operability. This dissertation proposes a personalized AI assessment module that discovers the high-level ``capabilities'' of an AI system with arbitrary internal planning algorithms/policies and learns an accurate symbolic description of these capabilities in terms of concepts that a user understands. Furthermore, the dissertation includes the associated theoretical results and the empirical evaluations. The results show that (i) a primitive query-response interface can enable the development of autonomous assessment modules that can derive a causally accurate user-interpretable model of the system's capabilities efficiently, and (ii) such descriptions are easier to understand and reason with for the users than the agent's primitive actions.

In the domain of speech recognition many methods have been proposed over time like Gaussian mixture models (GMM), GMM with universal background model (GMM-UBM framework), joint factor analysis, etc. i-Vector subspace modeling is one of the recent methods that has become the state of the art technique in this domain. This method largely provides the benefit of modeling both the intra-domain and inter-domain variabilities into the same low dimensional space. In this survey, we present a comprehensive collection of research work related to i-vectors since its inception. Some recent trends of using i-vectors in combination with other approaches are also discussed. The application of i-vectors in various fields of speech recognition, viz speaker, language, accent recognition, etc. is also presented. This paper should serve as a good starting point for anyone interested in working with i-vectors for speech processing in general. We then conclude the paper with a brief discussion on the future of i-vectors.

Non-Communicable Diseases (NCDs) are now a leading cause of mortality and morbidity globally, and mental illness is a significant part of it. In India, the treatment gap for common mental disorders is over 80%. In order to bridge this gap, mental health treatment models recommend task-shifting to non-specialists and integration of mental health care into general healthcare services. Other NCDs are being managed effectively by non-specialist healthcare workers (HCWs) at primary care, and mental illness and substance misuse are highly comorbid with other NCDs; hence, integrating mental health care within the NCD services and care framework seems logically feasible and effective. However, country-specific characteristics pose a significant challenge to the implementation of integrated care for mental disorders and NCDs. The primary objective of this study includes the development and implementation of a service delivery model that would result in at least 70% coverage of screening, linkage to care, and management of common mental disorders and substance use disorders (MSUD) among persons seeking care for NCDs at public health facilities. Secondary objectives include assessment of the feasibility of adoption of the implementation model by the health care system and to evaluate the cost of the mental health service strengthening intervention package from the health system's and the patient's perspectives. It will be a multi-site implementation research study, employing a mixed-methods quasi-experimental, within-site, three-phase, single-arm, interrupted time series design. The implementation model comprises screening, treatment, and linkage of mental health services integrated into NCD care in at least three blocks in each of the seven selected districts of the seven selected states of India, which are geographically far apart. The expected outcome would be to increase the proportion of patients screened and managed for MSUDs among persons seeking care for NCDs at the public health facilities. The results of this implementation research will provide a roadmap for scaling up of integrated MSUDs services within general healthcare.

The aim of this study was to analyze the role of thiol/disulfide homeostasis (TDH) parameters as an indicator of oxidative stress in acute appendicitis (AA). PubMed, EMBASE, Web of Science, and Scopus databases were systematically searched. Studies reporting on TDH in AA (both complicated and uncomplicated cases) were included. The comparator group were healthy controls. The TDH domain was compared between the groups using anti-oxidant parameters, namely native thiol and total thiol levels, and native thiol/total thiol ratio; and oxidant parameters, namely disulfide level, disulfide/native thiol ratio, and disulfide/total thiol ratio. The statistical analysis was performed using a random-effects model. The methodological quality of the studies was assessed utilizing the Newcastle–Ottawa scale. Eleven studies with a total of 926 subjects, comprising 457 patients with uncomplicated appendicitis, 147 with complicated appendicitis, and 322 healthy controls were included. Our study demonstrated significantly increased oxidative stress in AA as compared to healthy controls in all TDH parameters and significantly lower total thiol levels in complicated AA as compared to uncomplicated AA. Due to a poor methodological quality in five out of eleven studies, future prospective studies with adequate power are essential to validate these observations and refine the diagnostic approaches to AA.

This paper introduces a new approach for continual planning and model learning in relational, non-stationary stochastic environments. Such capabilities are essential for the deployment of sequential decision-making systems in the uncertain and constantly evolving real world. Working in such practical settings with unknown (and non-stationary) transition systems and changing tasks, the proposed framework models gaps in the agent's current state of knowledge and uses them to conduct focused, investigative explorations. Data collected using these explorations is used for learning generalizable probabilistic models for solving the current task despite continual changes in the environment dynamics. Empirical evaluations on several non-stationary benchmark domains show that this approach significantly outperforms planning and RL baselines in terms of sample complexity. Theoretical results show that the system exhibits desirable convergence properties when stationarity holds.

Reinforcement learning (RL) allows vision-language-action (VLA) policies to generalize beyond their training distribution by optimizing directly for task success, but post-training is computationally expensive. A natural response has been to speed rollout collection through faster simulators and world models. In GRPO-based VLA RL, we find that the dominant cost lies elsewhere: gradient computation accounts for approximately 78% of wall-clock time per step in our runs, while rollout collection accounts for only 21%. Gradient cost dominates because much of this computation is spent on phases that contribute little to learning. GRPO's learning signal is driven by advantage variance: only phases where successful and failed rollouts diverge produce learning signal. However, GRPO assigns the same advantage to every chunk in a rollout. As a result, actor-update compute is spent uniformly across the trajectory, including phases the policy already handles after pre-training and supervised fine-tuning. This paper presents Probabilistic Chunk Masking (PCM), a drop-in modification to GRPO that allocates gradient computation to a small, probabilistically selected subset of chunks per trajectory. PCM scores semantic phases using success-failure action variance, a rollout-derived proxy for per-phase gradient variance, and samples a fixed chunk budget with online-updated phase-level keep probabilities. We formalize per-phase gradient variance as the quantity determines where gradient computation is useful and show that success-failure action variance provides a measurable proxy for it. PCM requires no reward model or learned critic. On three LIBERO benchmarks, PCM matches the final success rate of standard GRPO while achieving 2.38 times wall-clock speedup, 4.8 times faster gradient updates, and 60% lower peak activation memory, while backpropagating through fewer than 20% of trajectory chunks.