India advances its clinical research capabilities through the adoption of artificial intelligence (AI). This development increases the speed and accuracy of clinical trials. It also strengthens data quality and supports better outcomes. This review outlines the current use of AI in Indian clinical trials. It presents key challenges and proposes solutions to improve future progress. A structured literature search covered databases such as PubMed and Scopus along with Google Scholar. This review focused on studies published between 2019 and 2025. AI provides clear advantages for clinical research. These include rapid patient recruitment, faster data validation, and improved safety monitoring. Tools such as automated systems and chatbot technologies reduce delays. They also enhance communication and support smoother trial processes. Several issues continue to affect progress in this area. Many regions lack reliable digital infrastructure. Data collected across trials often lack consistency. India also faces a shortage of professionals with expertise in both clinical research and AI. Ethical concerns continue to grow. These include limited safeguards for informed consent, weak data privacy, and potential bias within AI systems. To address these barriers, India must invest in better digital systems. Training programs should expand to prepare professionals for AI-integrated roles. Clear legal guidelines and ethical frameworks must be developed. Understanding should include members with a strong understanding of AI. With these reforms, India can establish global leadership in safe and effective use of AI in clinical research. This transformation will improve trial quality, enhance patient safety, and strengthen India’s international research reputation.
A calcifying odontogenic cyst (COC) is a rare developmental odontogenic cyst characterized by diverse clinical, radiographic, and histological features. This case series aims to highlight the variability in presentations and emphasize the importance of histopathological confirmation. Three cases of COC were evaluated using clinical examination, panoramic radiography, and histological analysis. All patients underwent surgical management and follow-up. Patients presented with painless swelling in different regions of the maxilla and mandible, consistent with the slow-growing nature of COC. Radiographically, the cyst appeared as well-defined, unilocular radiolucencies with varying degrees of calcifications. Histopathological examination revealed characteristic ghost cells and calcifications in all cases. Surgical enucleation resulted in satisfactory outcomes without recurrence during follow-up. This case series contributes to the understanding of COC and emphasizes the necessity of thorough clinical, radiographical, and histological evaluation for accurate diagnosis and appropriate management.
Background: Cataract surgery in patients with nondilating pupils poses significant intraoperative challenges and increases the risk of complications. Various mechanical pupil expansion techniques, including iris hooks, sphincterotomy, and pupil expanders, are used during phacoemulsification; however, comparative evidence regarding their early postoperative outcomes remains limited. The objective of the study is to compare the early postoperative efficacy and safety of iris hooks, sphincterotomy, and pupil expanders in patients undergoing phacoemulsification for nondilating pupils. Methods: This prospective observational cohort study included 55 patients with cataracts and nondilating pupils who underwent phacoemulsification using iris hooks (n = 20), sphincterotomy (n = 20), or pupil expanders (n = 15). Surgical outcomes, intraoperative complications, postoperative corneal clarity, pupil morphology, iris status, and best-corrected visual acuity were assessed during follow-up visits on day 1, week 1, and week 4. Statistical analysis was performed using appropriate parametric and nonparametric tests, with a significance level set at P < 0.05. Results: Baseline demographic characteristics were comparable across the three groups. Sphincterotomy achieved a 100% surgical success rate with no intraoperative complications, but resulted in permanent pupil irregularity in all cases. Iris hooks and pupil expanders preserved round pupil morphology in 80% and 60% of cases, respectively, but were associated with higher rates of minor intraoperative complications. Postoperative corneal clarity improved progressively in all groups, with complete resolution by week 4. Final visual acuity outcomes at week 4 were comparable among the groups, with no statistically significant difference observed (P > 0.05). Conclusion: All three techniques enable successful phacoemulsification in eyes with nondilating pupils during the early postoperative period. While sphincterotomy provides excellent intraoperative safety, it is associated with permanent pupil irregularity. Iris hooks and pupil expanders better preserve pupil morphology, but may carry a slightly higher risk of intraoperative complications. The choice of technique should be individualized based on intraoperative findings and surgical priorities. The short duration of follow-up is a limitation, and longer-term studies are warranted.
Background: Emergency departments (EDs) face increasing pressure from patient volumes, resource constraints, and time-critical decision-making demands. Artificial intelligence (AI), encompassing machine learning (ML), deep learning (DL), and natural language processing (NLP), has emerged as a potential tool to augment clinical decision support in this environment. However, evidence regarding its clinical effectiveness, safety, and implementation fidelity remains heterogeneous and incompletely synthesized. This systematic review aimed to critically evaluate AI-based clinical decision support systems in emergency medicine, with particular focus on discriminative performance, evidence of clinical impact, safety outcomes, and methodological quality. Methods: This review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 guidelines. Databases searched included PubMed/MEDLINE, Elsevier/ScienceDirect, Scopus, and the Cochrane Library through September 2025. Eligible studies included retrospective cohort studies, prospective studies, and AI model validation studies involving adult ED patients and evaluating ML, DL, or NLP algorithms for diagnosis, triage, risk stratification, or outcome prediction. Risk of bias was assessed using the Risk of Bias in Non-randomised Studies of Interventions tool, appropriate for the exclusively retrospective, nonrandomized evidence base. The review was not prospectively registered in PROSPERO, which is acknowledged as a methodological limitation. Results: Nine retrospective observational studies from six countries (USA, Singapore, Greece, Taiwan, Israel, and Australia) were included, encompassing over 1.5 million ED encounters. Studies were categorized as model development (n = 6), external or real-world validation (n = 2), or implementation/impact studies (n = 1). AI models consistently demonstrated superior discriminative performance over conventional scoring systems (Emergency Severity Index, Modified Early Warning Score, Sequential Organ Failure Assessment, and Systemic Inflammatory Response Syndrome), with area under the receiver operating characteristic curve (AUROC) values ranging from 0.80 to 0.97 for sepsis prediction, critical care triage, and risk stratification. Conclusion: NLP-enhanced models combining structured and unstructured data achieved the highest discrimination. Notably, however, improved AUROC did not uniformly translate into demonstrated clinical benefit. The commercially deployed Epic Sepsis Model exhibited substantially reduced performance in prerecognition settings (AUROC: 0.47), illustrating the risks of overfitting and poor external validity. Only one implementation study (Kotovich et al.) reported measurable reductions in 30-day and 120-day mortality following AI deployment for intracerebral hemorrhage detection, though this was subject to important methodological limitations. Formal safety endpoints were not prospectively measured in any included study. Risk of bias was assessed as moderate to high across the included literature. Conclusions: Current evidence supports the discriminative potential of AI-based clinical decision support in emergency medicine, particularly for early sepsis identification and triage prioritization. However, the evidence base is constrained by its exclusively retrospective design, absence of prospective safety measurement, limited external validation, and inadequate attention to calibration, algorithmic fairness, and workflow integration. AUROC improvements should not be conflated with clinical effectiveness. Prospective, multicenter impact trials measuring patient-centered outcomes, alongside rigorous evaluation of implementation fidelity, algorithmic fairness, and governance frameworks, are essential before widespread clinical adoption can be recommended.
Background: To date, limited data exist on analgesic utilization and its association with quality of life (QoL) among oncology patients. This study aims to assess the association between pain, analgesic use, and QoL of cancer patients. Methods: A descriptive cross-sectional survey was conducted among 408 cancer patients attending the outpatient oncology clinic at Lagos University Teaching Hospital. Sociodemographic data, types of cancer, types of treatment, analgesic utilization, description, and severity of pain were obtained using a well-structured self-designed proforma. QoL was assessed using the WHOQoL-BREF questionnaire. Regression analysis estimated the association between pain and QoL in an unadjusted/adjusted analysis. Results: The most common cancer in the female and male populations was breast cancer (51.7%, n = 218) and prostate cancer (5.9%, n = 25), respectively. The prevalence of analgesic utilization was 89.9%. The majority (70.1%, n = 286) of the participants had moderate-to-severe pain scores. Fifty-eight percent (n = 236) of participants had satisfactory health. QoL was rated as poor/very poor in 17% (n = 68). Compared to cancer patients with no to mild pain, those with moderate-to-severe pain had poorer QoL (odds ratio [OR] =0.12, 95% confidence interval [CI]: 0.04, 0.33). This relationship remained significant after adjusting for other covariates, including analgesic use (OR = 0.10, 95% CI: 0.03–0.29). Conclusion: A high prevalence of moderate to severe pain was observed among cancer patients, negatively affecting their QoL. Interestingly, this association was independent of analgesic use, suggesting that other pain management strategies are warranted to improve QoL.
