Category: Nevin Manimala

Neural Netw. 2026 Feb 26;200:108775. doi: 10.1016/j.neunet.2026.108775. Online ahead of print.

ABSTRACT

In this paper, we propose a single-qubit graph classifier that combines classical graph representation with quantum computing. Through a lightweight architecture, it achieves flexible and efficient graph data processing. This model delegates the task of aggregating node features that do not require inference to a classical subroutine, and optimizes the key weight training process with quantum programs, thereby constructing a basic graph classifier that uses only one qubit. Experimental results show that in binary classification tasks, the proposed single-qubit classifier demonstrates strong competitiveness in terms of performance when compared with traditional algorithms and quantum algorithms. Additionally, we utilize a parallel training scheme for multiple single-qubit classifiers to effectively enhance performance in multi-classification tasks. Evaluations conducted in different quantum noise simulation environments indicate that the proposed model has good robustness, and the parallel training scheme for multiple classifiers further enhances the model’s robustness in multi-classification tasks. More importantly, this model can be flexibly combined with various classical graph neural networks, providing a way to promote the application of quantum graph neural networks in diverse scenarios.

PMID:41791178 | DOI:10.1016/j.neunet.2026.108775

Neural Netw. 2026 Mar 1;200:108753. doi: 10.1016/j.neunet.2026.108753. Online ahead of print.

ABSTRACT

In this paper, we introduce a novel algorithm for image segmentation, which is based on the Allen-Cahn (AC) energy function. Our methodology involves calculating the energy feature at a local level in an image by means of the sliding window technique on the image matrix, which ultimately produces the energy matrix necessary for segmenting the image. Subsequently, constraints are constructed using the extreme values in the energy matrix. By varying the parameters in the constraints and the size of the sliding window, image segmentation results are obtained for different demand purposes. This paper empirically analyzes various simple and complex images, demonstrating the algorithm’s effectiveness in segmentation across different complexities and its excellent performance in agronomy and medicine. We also compare our method with other state-of-the-arts and our algorithm exhibits significant advantages in terms of time-saving. In order to further optimise the parameter selection process, we improve the algorithm so that it can autonomously output the parameters that make the segmentation results optimal. Additionally, we conduct experiments on several classical image segmentation datasets, as well as evaluating the segmentation results through the introduction of evaluation metrics, to demonstrate the effectiveness and applicability of our optimised algorithm. Our method demonstrates outstanding performance in image segmentation experiments on the CO-SKEL dataset, with an average segmentation time of 0.2 s. The accuracy exceeds 95%, and precision, recall, and F1 scores all surpass 90%. Even under various noise interferences, the segmentation accuracy of the algorithm remains above 94%, highlighting its efficiency and robustness.

PMID:41791175 | DOI:10.1016/j.neunet.2026.108753

Clin Oncol (R Coll Radiol). 2026 Feb 14;52:104084. doi: 10.1016/j.clon.2026.104084. Online ahead of print.

ABSTRACT

AIMS: Brain metastases (BMs) represent the most common intracranial tumours in adults with advanced solid cancers, significantly impacting morbidity and mortality. While the emergence of systemic therapies such as immunotherapy (IT) and targeted therapy (TT) has improved survival outcomes, the interaction of these modalities with stereotactic radiotherapy/radiosurgery (SRS/SRT) remains poorly explored. This study aims to evaluate the safety, local control (LC), and efficacy of combining SRS/SRT with IT or TT in patients with BM.

MATERIALS AND METHODS: This prospective, monocentric study (“RaBITT Trial”) analysed patients with BM treated with SRS/SRT using the CyberKnife system (Accuray) between May 2020 and May 2023. Patients received IT or TT in conjunction with SRS/SRT. Outcomes assessed included LC, overall survival (OS), disease-free survival (DFS), and next-line systemic treatment-free survival (NEST-FS). Toxicity, including radionecrosis and haemorrhage, was evaluated using the Common Terminology Criteria for Adverse Events (CTCAE) v5.0 criteria. Statistical analysis employed Kaplan-Meier curves and univariate (UVA) and multivariate (MVA) regression models.

RESULTS: We analysed 45 patients with a total of 225 BMs. The LC rate at 1 and 2 years was 85.7% and 79.8%, respectively. The overall response rate was 85.3%, with 56 complete responses (CRs). Multivariate analysis identified a higher biological effective dose with an α/β ratio of 10 and prescription isodose as predictors of CR. DFS rates at 1 and 2 years were 37.1% and 27.5%, respectively, while OS rates were 52.5% and 44.7%, respectively. NEST-FS rates at 1 and 2 years were 35.9% and 25.5%, respectively. Toxicity analysis revealed a radionecrosis rate of 20% for patients (6.7% of total lesions), with most cases being asymptomatic. Intralesional haemorrhage occurred in 1.7% of lesions, predominantly in melanoma and lung cancer patients.

CONCLUSION: The combination of SRS/SRT with IT or TT yields high rates of LC and overall response with a manageable safety profile, emphasising the importance of personalised dosimetric planning and careful toxicity monitoring. Despite promising outcomes, the study’s monocentric nature and limited sample size highlight the need for multicenter trials with extended follow-up to optimise treatment strategies and better understand long-term risks and benefits.

PMID:41791143 | DOI:10.1016/j.clon.2026.104084

J Neurosurg Spine. 2026 Mar 6:1-9. doi: 10.3171/2025.9.SPINE241288. Online ahead of print.

ABSTRACT

OBJECTIVE: Limited research has explored the impact of predominant back pain (pBP) versus predominant leg pain (pLP) in patients undergoing unilateral biportal endoscopic lumbar decompression (UBE-LD). This study aimed to evaluate and compare the perioperative and postoperative clinical outcomes, using patient-reported outcome measures (PROMs) and minimal clinically important difference (MCID), between patients with pBP and pLP who underwent UBE-LD.

METHODS: Patients who underwent primary UBE-LD were divided into either the pBP or pLP cohort. Exclusion criteria included patients with a diagnosis of degenerative scoliosis, trauma, malignancy, or infection. Demographic, perioperative characteristics, PROMs, and MCID were compared between cohorts using inferential statistics. PROMs were collected at preoperative and postoperative 6-week, 12-week, and 6-month time points. Assessed PROMs included the Patient-Reported Outcomes Measurement Information System-Physical Function (PROMIS-PF), Veterans Rand 12-Item Health Survey (VR-12) physical component score (PCS), VR-12 mental component score (MCS), visual analog scale (VAS) for back pain, VAS for leg pain, and Oswestry Disability Index (ODI). MCID attainment was determined by comparing the change in PROMs to established thresholds in the literature.

RESULTS: A total of 98 patients were included in the study, with 52 in the pBP cohort and 46 in pLP cohort. After analysis, there were no significant differences in baseline demographic or perioperative characteristics between the two groups. The majority of patients had a diagnosis of herniated nucleus pulposus (81.6%), central stenosis (93.9%), and foraminal stenosis (77.6%). At 6 months, both cohorts experienced significant postoperative improvements in PROMIS-PF, VAS back pain, VAS leg pain, VR-12 PCS, and ODI scores (all p ≤ 0.011). VR-12 MCS scores did not demonstrate sustained postoperative improvement in either cohort. High rates of MCID achievement were observed across both cohorts for multiple PROMs, with no statistically significant differences in MCID attainment between patients with pBP and pLP at any postoperative time point.

CONCLUSIONS: Regardless of the preoperative predominant pain location, patients undergoing UBE-LD reported significant improvements in physical function, back and leg pain, and disability. Patients in both cohorts demonstrated improvements in mental health outcomes, with comparable MCID achievement rates for VR-12 MCS. These findings suggest that UBE-LD might be effective for patients with both predominant back and leg pain presentations, although differences in specific outcome domains should be interpreted with caution due to potential selection bias. Consideration of predominant pain location might help inform preoperative patient counseling regarding expected outcomes following UBE-LD.

PMID:41791117 | DOI:10.3171/2025.9.SPINE241288

J Neurosurg. 2026 Mar 6:1-6. doi: 10.3171/2025.10.JNS251032. Online ahead of print.

ABSTRACT

OBJECTIVE: Patients experiencing ischemic strokes typically develop substantial cognitive decline. Intracranial atherosclerotic disease (ICAD) is a common stroke etiology that exposes patients to high and prolonged risks of recurrence. The ERSIAS-PC (Encephaloduroarteriosynangiosis revascularization for symptomatic intracranial atherosclerotic steno-occlusive performance criterion) phase II trial showed a lower risk of recurrent stroke in patients who underwent encephaloduroarteriosynangiosis (EDAS) plus intensive medical management (IMM). In the current study, the authors evaluate factors contributing to cognitive decline in patients with symptomatic ICAD treated with EDAS revascularization.

METHODS: ERSIAS-PC patients without aphasia who had completed at least 1 year of follow-up were included this post hoc analysis. Cognitive function was evaluated using the Montreal Cognitive Assessment (MoCA) at baseline and each follow-up and was classified as improved/preserved or worsened. Classification and regression tree (CART) analysis was used to identify factors associated with changes in cognitive function. The factors considered were age, sex, stenosis versus occlusion, baseline modified Rankin Scale score, good collateralization, and compliance with diabetes mellitus (DM), hypertension, and hyperlipidemia (HLD) treatments.

RESULTS: Of the 52 ERSIAS-PC patients, 39 were included in this subgroup analysis. The median age was 46 (IQR 37.0-56.0) years, and 27 (69.2%) patients were female. The mean MoCA score was 22.4 ± 4.9 at baseline and 23.9 ± 4.9 at the 1-year follow-up among the 52 patients in the ERSIAS-PC trial population. Among the 39 patients in this subgroup analysis, the MoCA score improved or remained stable in 33 (84.6%) and declined in 6 (15.4%). CART analysis indicated that the most relevant factor for an improved MoCA score after surgery was compliance with DM treatment (94.5% yes vs 74.2% no, p = 0.02). Other factors indicating a nominal though not statistically significant influence were HLD treatment (83.3% yes vs 60.5% no, p = 0.2) and stenosis (99.1% vs 80.9% occlusion, p = 0.6).

CONCLUSIONS: Compliance with DM treatment was significantly associated with cognitive preservation in patients with symptomatic ICAD treated with EDAS. The study findings emphasize the importance of the IMM of stroke risk factors in patients with intracranial atherosclerosis, even after surgical revascularization.

PMID:41791115 | DOI:10.3171/2025.10.JNS251032

JMIR Pediatr Parent. 2026 Mar 6;9:e74933. doi: 10.2196/74933.

ABSTRACT

BACKGROUND: Febrile seizures, although typically benign, can cause significant emotional distress for parents. Their diverse etiological risk factors underscore the need for further research. Ecological momentary assessment (EMA) offers a cost-effective and timely method for real-time data collection. The FeverApp, an EMA-based registry for fever management, enables parents to document febrile seizures as they occur.

OBJECTIVE: This study systematically investigates febrile seizure records from the FeverApp registry to assess their characteristics and explore the clinical implications of the findings. By providing real-world data on seizure management, this research demonstrates the potential of app-based EMA in pediatric care. Additionally, it offers insights for targeted interventions and improved febrile seizure management.

METHODS: We used a mixed methods approach. A descriptive qualitative analysis of parental descriptions of 226 seizures belonging to 161 children was conducted. Additionally, a comparative quantitative analysis of group differences was assessed through matched-pair sampling, comparing 114 children. Statistical methods were tailored to the nature of the respective variables, which included prevalence, age, gender, health and febrile history, fever management, temperature, well-being, and parental confidence.

RESULTS: Qualitative analyses provided detailed descriptions of seizure symptoms, seizure duration, and seizure management practices. Additionally, the data revealed a high rate of emergency consultations related to febrile seizures. However, there was underreporting of febrile seizures within the FeverApp, with a reported incidence of only 0.4% among febrile children. In a matched sample controlled for gender and age, significant differences were observed between febrile children with and those without febrile seizures in several parameters, including maximum recorded temperature (P<.001), prevalence of chronic diseases (P=.004), parental confidence (P=.01), and frequency of emergency consultations (P<.001).

CONCLUSIONS: This study offers valuable insights into the characteristics, temporal dynamics, management strategies, and parental responses to febrile seizures in children. Despite the limitation of potential underreporting in an EMA-based registry, the findings highlight the critical importance of parental education and support in managing febrile seizures. Enhancing these areas has the potential to reduce unnecessary medical consultations and improve the overall care of affected children. Furthermore, integrating improvements in the FeverApp’s education and documentation system regarding febrile seizures could facilitate better management and support future research efforts.

PMID:41791112 | DOI:10.2196/74933

JMIR Mhealth Uhealth. 2026 Mar 6;14:e77935. doi: 10.2196/77935.

ABSTRACT

BACKGROUND: Accurate limb circumference measurement is essential for the diagnosis and management of lymphedema, facilitating the objective evaluation of treatment outcomes and early detection of disease progression. Although manual tape measurements are inexpensive and widely used, they are limited by interobserver variability and low reproducibility. Advanced modalities such as computed tomography and magnetic resonance imaging provide high precision but are costly and impractical for routine or home-based use. To overcome these limitations, we developed and validated a novel 3D measurement system that integrates a marker-based bodysuit (ZOZOSUIT2) with a smartphone app, enabling fast, accurate, and reproducible limb circumference assessment.

OBJECTIVE: This study aimed to evaluate the accuracy, repeatability, and time efficiency of the 3D measurement bodysuit and smartphone system compared with those of manual tape measurements in patients with upper and lower limb lymphedema.

METHODS: This prospective study included 26 patients (n=10 with upper limb lymphedema and n=16 with lower limb lymphedema). Using the ZOZOSUIT2, which contains approximately 20,000 dot markers, 12 full-body photographs were captured with a smartphone, and a 3D model was automatically generated. Circumference values were extracted at 10 standardized points on the upper limb (from the wrist to the top of the upper arm) and 9 on the lower limb (from the ankle to the base of the thigh). Each measurement was repeated 3 times to assess repeatability. Manual tape measurements were used as the reference standard. For each measurement point, the repeatability (intrameasurement variation) and the mean absolute error between the 2 methods were calculated. Measurement time was also recorded for both methods and compared statistically using the Wilcoxon signed-rank test.

RESULTS: Across all measurement points, the median repeatability of the bodysuit smartphone system ranged from 2.2 to 6.4 mm, indicating high reproducibility. When compared with manual measurements, the median of the mean absolute error for upper limb points ranged from 6.7 to 20.5 mm, and for lower limb points ranged from 5.5 to 15.9 mm. Relatively larger discrepancies were observed at the wrist (median 19.8, IQR 15.5-24.1 mm), 5 cm proximal to the wrist (median 20.5, IQR 8.5-28.1 mm), and the base of the thigh (median 15.9, IQR 10.2-25.9 mm). Mean measurement time was significantly shorter with the bodysuit smartphone system (88.2, SD 16.0 seconds) than with manual tape measurement (293.8, SD 58.5 seconds; P<.001).

CONCLUSIONS: The bodysuit smartphone system enables rapid, precise, and highly reproducible limb circumference assessment in patients with lymphedema. Despite minor differences at anatomical sites with complex contours, such as the wrist and the thigh base, the overall accuracy and time efficiency were clinically acceptable. This system may serve as a practical and scalable solution for both clinical and home-based lymphedema monitoring, contributing to the objective and standardized assessment of limb volume.

PMID:41791111 | DOI:10.2196/77935

J Med Internet Res. 2026 Mar 6;28:e85516. doi: 10.2196/85516.

ABSTRACT

BACKGROUND: Surveys show that many people are willing to use generative artificial intelligence (AI) for health questions. Prior research has largely focused on chatbot accuracy, with some studies finding that both physicians and consumers overwhelmingly prefer chatbot-generated text over physician responses.

OBJECTIVE: This study aimed to characterize and compare the emotional content of responses from physicians and 2 AI chatbots (OpenAI’s ChatGPT and Google’s Gemini) and to assess differences in reading level and use of medical disclaimers.

METHODS: A public, patient-deidentified telehealth website was used to compile 100 physician-answered questions. The same questions were posed to both chatbots between May 18 and 19, 2025. Two coders classified the emotional content of each sentence using a predefined codebook and reviewed for agreement. Emotions were ranked as primary, secondary, and tertiary by the proportion of sentences classified as each emotion per response. Multinomial logistic regression compared emotional rankings using physician responses as the reference. Word count, Flesch Reading Ease, and Flesch-Kincaid Grade Level were analyzed via ANOVA with the Tukey honestly significant difference test. Disclaimer use was compared between chatbots using a χ² test.

RESULTS: Primary emotions were overwhelmingly neutral, except for one response from each chatbot in which anger was primary. For secondary emotions, the odds ratio of hope was 80.28% (95% CI 37.71%-93.76%) lower for ChatGPT, while the odds ratio of fear was 3.29 (95% CI 1.44-7.49) times higher for Gemini. For tertiary emotions, the odds ratio of compassion was 1.94 (95% CI 1.06-3.54) times higher, and the odds ratio of having no tertiary emotion was 84.33% (95% CI 64.72%-93.04%) lower for Gemini. Gemini responses averaged 889.1 (SD 305.7) words, ChatGPT 476.5 (SD 109.5), and physicians 193.5 (SD 113.6). Gemini had the lowest average Flesch Reading Ease score at 39.9 (SD 8.8), followed by ChatGPT at 45.8 (SD 12.8), while physicians had the highest at 51.9 (SD 13.6). Gemini had the highest average Flesch-Kincaid Grade Level at 11.3 (SD 1.5), followed by ChatGPT at 9.9 (SD 1.9), and physicians at 9.2 (SD 2.4). Gemini was significantly more likely to include a disclaimer than ChatGPT (χ²₁=49.2; P<.001).

CONCLUSIONS: Chatbot responses were significantly (P<.001) longer and more difficult to read than physician responses and were more likely to contain a wider range of emotions. Qualitatively, chatbot responses were more varied in their presentation as well as in the breadth of the emotions themselves. The findings of this study could be used to inform more emotionally connected physician responses to patient message queries.

PMID:41791109 | DOI:10.2196/85516

JMIR Res Protoc. 2026 Mar 6;15:e84038. doi: 10.2196/84038.

ABSTRACT

BACKGROUND: Areca nut (AN) is a commonly consumed psychoactive substance, especially in South and Southeast Asia. Chronic chewing of AN has been linked to multiple health problems, including temporomandibular joint (TMJ) disorders. Excessive strain on TMJ during chronic AN chewing can lead to repetitive injury, resulting in microtrauma and macrotrauma to both the TMJ and the surrounding masticatory structures. Previous studies have reported the long-term impact of AN chewing on TMJ by using conventional 2D imaging.

OBJECTIVE: This study aims to evaluate and compare condylar morphology in chronic AN chewers and nonchewers by using 3D imaging.

METHODS: This study will include 90 patients who will be divided into 2 groups: chronic AN chewers (n=45, 50%) and nonchewers (n=45, 50%). The study will be undertaken after obtaining institutional ethics committee approval and written informed consent from each patient. A detailed habit history of all the participants will be recorded. Each patient will undergo a clinical examination and radiographic evaluation of condylar morphology. Condylar morphology will be evaluated using cone beam computed tomography scans in both sagittal and coronal planes. All the findings will be recorded and then examined for statistical significance.

RESULTS: On comparison of condylar morphology between chronic AN chewers and nonchewers by using cone beam computed tomography, statistical variations relevant to structural and pathological alterations such as osteophytes, surface flattening, and erosions are likely to occur.

CONCLUSIONS: This study aims to overcome the limitations of conventional 2D radiography and provide a more accurate assessment of condylar morphology. The findings should fill an existing gap in the literature by providing useful insights on the effects of chronic AN chewing on condylar structure by using 3D imaging. This research may help to improve the diagnosis, prevention, and management of TMJ disorders.

TRIAL REGISTRATION: Clinical Trials Registry-India CTRI/2025/06/088238; https://ctri.nic.in/Clinicaltrials/pmaindet2.php?EncHid=MTMzNzQz&Enc=&userName=.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): PRR1-10.2196/84038.

PMID:41791104 | DOI:10.2196/84038

Phys Rev Lett. 2026 Feb 20;136(7):077101. doi: 10.1103/xgkj-dxzh.

ABSTRACT

We propose a method for inferring entropy production (EP) in high-dimensional stochastic systems, including many-body systems and non-Markovian systems with long memory. Standard techniques for estimating EP become intractable in such systems due to computational and statistical limitations. We infer trajectory-level EP and lower bounds on average EP by exploiting a nonequilibrium analogue of the maximum entropy principle, along with convex duality. Our approach uses only samples of trajectory observables, such as spatiotemporal correlations. It does not require reconstruction of high-dimensional probability distributions or rate matrices, nor impose any special assumptions such as discrete states or multipartite dynamics. In addition, it may be used to compute a hierarchical decomposition of EP, reflecting contributions from different interaction orders, and it has an intuitive physical interpretation as a “thermodynamic uncertainty relation.” We demonstrate its numerical performance on a disordered nonequilibrium spin model with 1000 spins and a large neural spike-train dataset.

PMID:41791056 | DOI:10.1103/xgkj-dxzh