Nevin Manimala

JMIR Cancer. 2025 Sep 23;11:e71102. doi: 10.2196/71102.

ABSTRACT

BACKGROUND: Complementary therapies are being increasingly used by cancer survivors. As a channel for customers to share their feelings, outcomes, and perceived knowledge about the products purchased from e-commerce platforms, Amazon consumer reviews are a valuable real-world data source for understanding cancer survivorship care needs.

OBJECTIVE: In this study, we aimed to highlight the potential of using Amazon consumer reviews as a novel source for identifying cancer survivorship care needs, particularly related to symptom self-management. Specifically, we present a publicly available, manually annotated corpus derived from Amazon reviews of health-related products and develop baseline natural language processing models using deep learning and large language model (LLM) to demonstrate the usability of this dataset.

METHODS: We preprocessed the Amazon review dataset to identify sentences with cancer mentions through a rule-based method and conducted content analysis including text feature analysis, sentiment analysis, topic modeling, cancer type, and symptom association analysis. We then designed an annotation guideline, targeting survivorship-relevant constructs. A total of 159 reviews were annotated, and baseline models were developed based on deep learning and large language model (LLM) for named entity recognition and text classification tasks.

RESULTS: A total of 4703 sentences containing positive cancer mentions were identified, drawn from 3349 reviews associated with 2589 distinct products. The identified topics through topic modeling revealed meaningful insights into cancer symptom management and survivorship experiences. Examples included discussions of green tea use during chemotherapy, cancer prevention strategies, and product recommendations for breast cancer. Top 15 symptoms in reviews were also identified, with pain being the most frequent symptom, followed by inflammation, fatigue, etc. The annotation labels were designed to capture cancer types, indicated symptoms, and symptom management outcomes. The resulting annotation corpus contains 2067 labels from 159 Amazon reviews. It is publicly accessible, together with the annotation guideline through the Open Health Natural Language Processing (OHNLP) GitHub. Our baseline model, Bert-base-cased, achieved the highest weighted average F1-score, that is, 66.92%, for named entity recognition, and LLM gpt4-1106-preview-chat achieved the highest F1-score for text classification tasks, that is, 66.67% for “Harmful outcome,” 88.46% for “Favorable outcome” and 73.33% for “Ambiguous outcome.”

CONCLUSIONS: Our results demonstrate the potential of Amazon consumer reviews as a novel data source for identifying persistent symptoms, concerns, and self-management strategies among cancer survivors. This corpus, along with the baseline natural language processing models developed for named entity recognition and text classification, lays the groundwork for future methodological advancements in cancer survivorship research. Importantly, insights from this study could be evaluated against established clinical guidelines for symptom management in cancer survivorship care. By revealing the feasibility of using consumer-generated data for mining survivorship-related experiences, this study offers a promising foundation for future research and argumentation analysis aimed at improving long-term outcomes and support for cancer survivors.

PMID:40986859 | DOI:10.2196/71102

JMIR Form Res. 2025 Sep 23;9:e69394. doi: 10.2196/69394.

ABSTRACT

BACKGROUND: In the United Kingdom, access to and the quality of palliative and end-of-life care (PEOLC) vary widely. In the final months of life, many patients face avoidable accident and emergency (A&E) visits and hospital admissions, driven by gaps in out-of-hours support and poorly coordinated care. This not only increases stress for patients and carers but also places avoidable strain and cost on the National Health Service (NHS). There is an urgent need for more compassionate, person-centered models that support people to remain at home, improve their quality of life (QoL), and reduce unnecessary use of acute services.

OBJECTIVE: This study aimed to explore the usability, user experiences, and impact of the digital dyadic remote monitoring Care and Support System for Patients and Carers (CARE-PAC) for patients in the last year of life, their informal carers, and health professionals involved in their care.

METHODS: Patients and informal carers were recruited to use CARE-PAC for up to 12 weeks. A mixed methods approach was used. Quantitative methods included the use of validated QoL scales and the System Usability Scale (SUS). Paired QoL and usability data were analyzed using the Wilcoxon (Pratt) signed-rank test, while unpaired usability data were analyzed using the Wilcoxon rank-sum test. Qualitative methods involved short catch-up calls, in-depth interviews, and focus groups conducted using topic guides informed by the domains of the Non-adoption, Abandonment, Scale-up, Spread, and Sustainability (NASSS) framework. Data were analyzed thematically.

RESULTS: CARE-PAC was implemented across 5 UK clinical sites with 26 participants (13 patient-carer dyads). No significant changes were observed in patients’ total QoL scores; however, significant improvements were seen in the “overall QoL” and “social” domains, alongside a significant decline in the “physical” domain. Carers showed no significant changes across total or domain-specific QoL scores. Usability was rated highly by patients (mean 87.9, SD 12.4) and carers (mean 94.7, SD 3.8), indicating an excellent user experience. Health care professionals (HCPs) reported lower usability scores (mean 63.6, SD 15.6), falling below average but above the threshold for poor usability. Thematic analysis of qualitative data gathered via catch-up calls (all patient-carer dyads), in-depth interviews (2 patients-2 carers), and 4 focus groups/1 interview (12 HCPs) identified 4 key themes: impact on care experiences, reflections and satisfaction, implementation challenges, and future directions.

CONCLUSIONS: CARE-PAC is a usable, feasible, and acceptable remote monitoring and support system for patients in the last year of life, their carers, and HCPs. It enables real-time identification of needs and has shown positive impacts on the QoL of both patients and carers. These findings support the need for further research to evaluate its effectiveness at scale and explore pathways for wider implementation in PEOLC.

PMID:40986856 | DOI:10.2196/69394

J Med Internet Res. 2025 Sep 23;27:e72051. doi: 10.2196/72051.

ABSTRACT

BACKGROUND: The effects of telemedicine on health expenditures and health outcomes are an important policy question. Many countries loosened regulations on the use of telemedicine during the COVID-19 pandemic, thereby offering an opportunity to evaluate these effects via a natural experiment.

OBJECTIVE: This study aimed to assess the effect of greater telemedicine use on area-level health expenditures and health outcomes related to common chronic conditions in Japan during the COVID-19 pandemic.

METHODS: We compared prefectures (area levels of government) with higher prepandemic telemedicine rates (fiscal year [FY] 2019) versus those with lower rates and conducted a difference-in-differences analysis of the change in prefecture-level health expenditures from FY2017 to FY2022 and health outcomes from FY2017 to FY2021. The participants were the total population in Japan from FY2017 to FY2022 (n=126 million), and the exposure was the increase in telemedicine use following the government’s relaxation of restrictions on telemedicine use as an exceptional measure during the COVID-19 pandemic. Our main outcomes were the share of outpatient claims that were for telehealth services; total, inpatient, and outpatient annual prefecture-level health expenditures; all-cause mortality, glycated hemoglobin, systolic blood pressure, and low-density lipoprotein cholesterol.

RESULTS: Treatment prefectures (n=15, population of 62 million) were defined as those with greater-than-median telemedicine use before the pandemic, while control prefectures (n=32, population of 64 million) were defined as those with less-than-median telemedicine use. Treatment and control prefectures shared similar demographic characteristics before the pandemic. The growth in telemedicine after 2020 as a share of outpatient claims increased among the treatment prefectures by 0.35 percentage points more than among control prefectures, which represented more than a threefold increase in telemedicine use compared to the prepandemic median. In difference-in-differences analyses, this difference was associated with a 1.0% relative decrease (95% CI 0.3%-1.8%) in total health expenditure (P=.006) and a 1.1% relative decrease (95% CI 0.2%-2.0%) in inpatient expenditure (P=.02). Outpatient expenditures showed no significant difference as a result of increased telemedicine adoption. Most health outcomes-all-cause mortality, glycated hemoglobin, systolic blood pressure, diastolic blood pressure, and low-density lipoprotein cholesterol-did not show any significant changes.

CONCLUSIONS: Areas in Japan with greater expansion of telemedicine use during the pandemic experienced a significant decrease in both inpatient and total health care spending compared with areas with less telemedicine use, without harming health outcomes.

PMID:40986854 | DOI:10.2196/72051

JMIR Aging. 2025 Sep 23;8:e73074. doi: 10.2196/73074.

ABSTRACT

BACKGROUND: Mild cognitive impairment (MCI) is a precursor of dementia. Therefore, MCI identification and monitoring are crucial to delaying dementia onset. Given the limits of existing clinical tests, objective support tools are needed.

OBJECTIVE: This work investigates quantitative handwriting analysis, tailored to enable domestic monitoring, as a noninvasive approach for MCI screening and assessment.

METHODS: A sensorized ink pen, used on paper and equipped with sensors, memory, and a communication unit, was used for data acquisition. The tasks included writing a grocery list and free text to mimic daily life handwriting, and a clinical dictation test (parole-non-parole [PnP] test), featuring regular, irregular, and made-up words, aimed at assessing MCI dysgraphia. From the recorded data, 106 indicators describing the performance in terms of time, fluency, exerted force, and pen inclination were computed. A total of 57 patients with MCI were recruited, of whom 45 performed a test-retest protocol. The indicators were examined to assess their test-retest reliability. The indicators from the test repetition were used to assess their relationship with the scores of clinical tests via correlation analysis. For the PnP test, differences in the indicators among the 3 types of words were statistically investigated. These analyses were conducted separately for the cursive (2/3 of the sample) and block letters (1/3 of the sample) allographs, with the level of significance set at 5%. Data from healthy older adults were available for the grocery list (34 participants) and free text (45 participants) tasks. These were exploited to build machine learning classification models for the distinction between patients with MCI and healthy controls.

RESULTS: When dealing with reliability, 93% and 44% of the indicators were characterized by a significant reliability of at least moderate intensity for cursive and block letters respectively. As for the correlation analysis, patients with preserved cognitive status and daily life functionality were associated with significantly better temporal performances, both in free writing and PnP. The analysis of PnP highlighted the presence of surface dysgraphia in the recruited sample, as irregular words showed significantly worse temporal indicators with respect to regular and made-up ones. The classification models’ built-in free writing data achieved accuracies ranging from 0.80 to 0.93 and F₁-scores from 0.81 to 0.92 according to the input dataset.

CONCLUSIONS: The presented results suggest the suitability of ecological handwriting analysis for the all-around monitoring of MCI, from early screening to disease progression evaluation.

PMID:40986851 | DOI:10.2196/73074

JMIR Form Res. 2025 Sep 23;9:e66237. doi: 10.2196/66237.

ABSTRACT

BACKGROUND: The global outbreak of COVID-19 has significantly impacted health care systems and has necessitated timely access to information for effective decision-making by health care authorities. Conventional methods for collecting patient data and analyzing virus mutations are resource-intensive. In the current era of rapid internet development, information on COVID-19 infections could be collected by a novel approach that leverages social media, particularly Twitter (subsequently rebranded X).

OBJECTIVE: The aim of this study was to analyze the trending patterns of tweets containing information about various COVID-19 symptoms, explore their synchronization and correlation with conventional monitoring data, and provide insights into the evolution of the virus. We categorized tweet sentiments to understand the predictive power of negative emotions of different symptoms in anticipating the emergence of new Omicron subvariants and offering real-time assistance to affected individuals.

METHODS: Relevant user tweets from 2022 containing information about COVID-19 symptoms were extracted from Twitter. Our fine-tuned RoBERTa model for sentiment analysis, achieving 99.7% accuracy for sentiment analysis, was used to categorize tweets as negative, positive, or neutral. Joinpoint regression analysis was used to examine the trends in weekly negative tweets related to COVID-19 symptoms, aligning these trends with the transition periods of SARS-CoV-2 Omicron subvariants from 2022. Real-time Twitter users with negative sentiments were geographically plotted. A total of 105,934 tweets related to fever, 120,257 to cough, 55,790 to headache, 101,220 to sore throat, 3410 to vomiting, and 5913 to diarrhea were collected.

RESULTS: The most prominent topics of discussion were fever, sore throat, and headache. The weekly average daily tweets exhibited different fluctuation patterns in different stages of subvariants. Specifically, fever-related negative tweets were more sensitive to Omicron subvariant evolution, while discussions of other symptoms declined and stabilized following the emergence of the BA.2 variant. Negative discussions about fever rose to nearly 40% at the beginning of 2022 and showed 2 distinct peaks during the absolute dominance of BA.2 and BA.5, respectively. Headache and throat-related negative sentiment exhibited the highest levels among the analyzed symptoms. Tweets containing geographic information accounted for 1.5% (1351/391,508) of all collected data, with negative sentiment users making up 0.35% (5873/391,508) of all related tweets.

CONCLUSIONS: This study underscores the potential of using social media, particularly tweet trends, for real-time analysis of COVID-19 infections and has demonstrated correlations with major symptoms. The degree of negative emotions expressed in tweets is valuable in predicting the emergence of new Omicron subvariants of COVID-19 and facilitating the provision of timely assistance to affected individuals.

PMID:40986849 | DOI:10.2196/66237

West Afr J Med. 2025 Apr 30;42(4):290-297.

ABSTRACT

BACKGROUND AND OBJECTIVES: Despite the burden and severity of AKI in ICU patients, there is limited epidemiologic information in Sub-Saharan Africa. Epidemiologic data on AKI in critically ill patients are required to advocate for government health policies that will reduce the burden of AKI. This study determined the incidence, aetiologies, and short-term outcomes of AKI in ICU patients in Southern Nigeria.

METHODS: A multi-centre retrospective descriptive study of all patients with medical conditions, who developed AKI in the ICU of the participating hospitals during the study period. Data collected include socio-demographics, hospital-related data such as duration of stay in ICU, past and current medical history, and outcome of illness.

RESULTS: Out of 473 cases, AKI was diagnosed in 203 (42.9%). Seventy-three (36.0%) were oliguric, 50.2% had non-oliguric AKI, while it was not specified in 13.8% patients. The most common causes of AKI were hypovolaemia (34.0%) and sepsis (27.6%). Factors associated with the development of AKI were sepsis [AOR:6.17; CI:2.33-16.32; p = <0.001] and need for inotropes [AOR:2.03; CI:1.34-3.94; P = 0.003]. Although dialysis was indicated in 34 (16.7%) of the AKI patients, 58.8% of them received it. One hundred and sixteen (57.1%) of the AKI patients were discharged with full renal recovery, while 40.9% died.

CONCLUSION: Four out of every 10 patients admitted into the ICU developed AKI, and the common aetiologies were hypovolaemia and sepsis. Regular critical care training is required to effectively identify at-risk patients and to take prompt measures towards mitigating these risks.

PMID:40986848

J Craniofac Surg. 2025 Sep 23. doi: 10.1097/SCS.0000000000011775. Online ahead of print.

ABSTRACT

INTRODUCTION: Cranial defects, often resulting from decompressive craniectomy, require timely repair to mitigate neurological risks and psychological sequelae. This study compares polyetheretherketone (PEEK) and autologous bone grafts in cranial reconstruction, focusing on clinical outcomes, cost-effectiveness, and utilization trends.

METHODS: A retrospective analysis was conducted on the clinical data of 188 patients who underwent skull repair surgery in the neurosurgery department of the First Affiliated Hospital of Shandong First Medical University (Shandong Provincial Qianfoshan Hospital) from June 1, 2015, to August 31, 2022. The patients were divided into a polyetheretherketone (PEEK) group (n=80) and an autogenous bone flap group (n=108) based on the different materials used for skull repair. Analyze the baseline data, repair materials, skull defect duration, hospitalization expenses, surgical information (duration of surgery, presence or absence of blood transfusion), and postoperative complications (epilepsy, hydrocephalus, subdural effusion, bone resorption, and infection) of the 2 groups of patients.

RESULTS: There was no significant difference in baseline data, primary disease, and time of skull defect between the 2 groups of patients (P>0.05). Comparison showed that traumatic brain injury (54.80%) was the main cause of skull repair surgery, followed by bleeding (22.30%) and cerebral infarction (18.10%). There was no significant difference in hospitalization time between the 2 groups of patients (P>0.05), and there was no significant difference in the postoperative recovery of skull repair materials. However, the hospitalization cost of the PEEK group was higher than that of the autogenous bone group, and the difference was statistically significant (P<0.05). The surgery time of PEEK group patients was shorter than that of the autogenous bone group, with PEEK 144 (109-185.75) minutes and autogenous bone 171.5 (155.25-187) minutes, respectively, and the difference was statistically significant (P<0.05). In terms of postoperative complications, a total of 82 people experienced complications during the 3-month follow-up, including 19 people in the PEEK group (23.70%) and 63 people in the autogenous bone group (58.30%). There was no significant difference between the 2 groups in postoperative epilepsy, hydrocephalus, infection, etc. (P>0.05). In terms of material selection for repairs, the use of PEEK has been on the rise year by year, and as the number of repair personnel tends to stabilize, the proportion of PEEK usage has been increasing year by year.

CONCLUSION: PEEK emerges as a cost-effective, clinically robust alternative to autologous bone grafts, particularly for complex defects. Policy interventions, including DRG-exempt payments and volume-based procurement, could further enhance its accessibility. These findings support broader adoption of PEEK in cranial reconstruction, balancing upfront costs with long-term clinical and economic benefits.

PMID:40986847 | DOI:10.1097/SCS.0000000000011775

J Craniofac Surg. 2025 Sep 23. doi: 10.1097/SCS.0000000000011741. Online ahead of print.

ABSTRACT

BACKGROUND: Metopic craniosynostosis (MCS) presents unique diagnostic challenges. Objective measures to correctly diagnose MCS have been developed, but mostly rely on thresholds obtained through subjective clinical diagnosis. This study quantifies the spectrum of normal infant anterior cranial shapes using an entirely automated method to aid in diagnosis of MCS using population statistics.

METHODS: A retrospective review of computed tomography (CT) scans from patients aged 0 to 24 months was completed. Patients with known craniosynostosis or associated syndromes, hydrocephalus, or other brain or cranial abnormalities were excluded. Optimal interfrontal angle (oIFA), transverse forehead width (TFW), skull circumference (SC), intracranial volume (ICV), and malformation range (MR) were calculated using an automated analysis pipeline.

RESULTS: A total of 582 subjects were included. Median age at CT scan was 11.6 months. Males demonstrated significantly higher median ICV, TFW, and SC than females (P<0.001 for all). OIFA was inversely correlated with age, decreasing by approximately 8 degrees between neonates and 24-month-old children. Transverse forehead width, SC, and ICV all increased significantly over the same age range (P<0.001 for all).

CONCLUSION: This study provides the largest and most comprehensive evaluation of normal frontal cranial shape in infants to date. The data presented show that forehead convexity follows a normal distribution after correcting for age. These oIFA measurements were collected using an automated method, allowing analysis of forehead shape without any basis in subjective interpretation. This data provides a basis for abandoning subjective, clinical diagnosis of MCS.

PMID:40986846 | DOI:10.1097/SCS.0000000000011741

J Craniofac Surg. 2025 Sep 23. doi: 10.1097/SCS.0000000000011976. Online ahead of print.

ABSTRACT

The impact of psychiatric disorders on oral/maxillofacial diseases through sleep apnea syndrome (SAS) remains incompletely understood. Using bidirectional and multivariable Mendelian randomization (MR), this study aimed to investigate potential causal links between psychiatric disorders, SAS, and oral/maxillofacial diseases, while assessing the mediating role of SAS. The authors analyzed genome-wide association study (GWAS) summary statistics using univariable MR to evaluate whether genetically predicted psychiatric disorders influence oral/maxillofacial manifestations; bidirectional MR and mediation MR were used to determine causal directionality and mediation effects. Univariable MR revealed that major depressive disorder increased risks of dentofacial anomalies (OR=1.25, 95% CI: 1.06-1.48, P=0.007), temporomandibular disorders (TMD) (OR=1.54, P=8×10-5), and temporomandibular muscle pain (OR=1.52, P=0.0008); post-traumatic stress disorder elevated risks of dentofacial anomalies (OR=1.07, P=0.02) and TMD (OR=1.09, P=0.04); autism spectrum disorder was associated with temporomandibular muscle pain (OR=8.10, P=0.008). Bidirectional MR confirmed mutual causation between SAS and dentofacial anomalies. Mediation analysis estimated SAS mediated 21.5% (95% CI: 13.1%-31.2%) of the effect of major depressive disorder on dentofacial anomalies. Psychiatric disorders exert causal effects on oral/maxillofacial diseases, partially mediated by SAS; these results highlight SAS as a mediator between psychiatric disorders and oral/maxillofacial diseases and underscore its bidirectional causality with dentofacial anomalies, suggesting novel targets for preventive interventions.

PMID:40986834 | DOI:10.1097/SCS.0000000000011976

Proc Natl Acad Sci U S A. 2025 Sep 30;122(39):e2509444122. doi: 10.1073/pnas.2509444122. Epub 2025 Sep 23.

ABSTRACT

The phenotypic effect of a mutation may depend on the genetic background in which it occurs, a phenomenon referred to as epistasis. One source of epistasis in proteins is direct interactions between residues in close physical proximity to one another. However, epistasis may also occur in the absence of specific interactions between amino acids if the genotype-to-phenotype map is nonlinear. Disentangling the contributions of these two phenomena-specific and global epistasis-from noisy, high-throughput mutagenesis experiments is highly nontrivial: The form of the nonlinearity is generally not known and model misspecification may lead to over- or underestimation of specific epistasis. In contrast to previous approaches, we do not attempt to model the fitness measurements directly. Rather, we begin with the observation that global epistasis, under the assumption of monotonicity, imposes strong constraints on the rank statistics of a combinatorial mutagenesis experiment. Namely, the rank-order of mutant phenotypes should be preserved across genetic backgrounds. We exploit this constraint to devise a simple semiparametric method to detect specific epistasis in the presence of global epistasis and measurement noise. We apply this method to three high-throughput mutagenesis experiments, uncovering known protein contacts with similar accuracy to existing, more complicated procedures. Our method immediately generalizes beyond proteins, providing a simple, yet powerful framework for interpreting the epistasis observed in combinatorial datasets.

PMID:40986352 | DOI:10.1073/pnas.2509444122