Tag: pubmed

J Med Internet Res. 2025 Jan 6;27:e66220. doi: 10.2196/66220.

ABSTRACT

BACKGROUND: The increasing use of social media to share lived and living experiences of substance use presents a unique opportunity to obtain information on side effects, use patterns, and opinions on novel psychoactive substances. However, due to the large volume of data, obtaining useful insights through natural language processing technologies such as large language models is challenging.

OBJECTIVE: This paper aims to develop a retrieval-augmented generation (RAG) architecture for medical question answering pertaining to clinicians’ queries on emerging issues associated with health-related topics, using user-generated medical information on social media.

METHODS: We proposed a two-layer RAG framework for query-focused answer generation and evaluated a proof of concept for the framework in the context of query-focused summary generation from social media forums, focusing on emerging drug-related information. Our modular framework generates individual summaries followed by an aggregated summary to answer medical queries from large amounts of user-generated social media data in an efficient manner. We compared the performance of a quantized large language model (Nous-Hermes-2-7B-DPO), deployable in low-resource settings, with GPT-4. For this proof-of-concept study, we used user-generated data from Reddit to answer clinicians’ questions on the use of xylazine and ketamine.

RESULTS: Our framework achieves comparable median scores in terms of relevance, length, hallucination, coverage, and coherence when evaluated using GPT-4 and Nous-Hermes-2-7B-DPO, evaluated for 20 queries with 76 samples. There was no statistically significant difference between GPT-4 and Nous-Hermes-2-7B-DPO for coverage (Mann-Whitney U=733.0; n₁=37; n₂=39; P=.89 two-tailed), coherence (U=670.0; n₁=37; n₂=39; P=.49 two-tailed), relevance (U=662.0; n₁=37; n₂=39; P=.15 two-tailed), length (U=672.0; n₁=37; n₂=39; P=.55 two-tailed), and hallucination (U=859.0; n₁=37; n₂=39; P=.01 two-tailed). A statistically significant difference was noted for the Coleman-Liau Index (U=307.5; n₁=20; n₂=16; P<.001 two-tailed).

CONCLUSIONS: Our RAG framework can effectively answer medical questions about targeted topics and can be deployed in resource-constrained settings.

PMID:39761554 | DOI:10.2196/66220

J Med Internet Res. 2025 Jan 6;27:e52601. doi: 10.2196/52601.

ABSTRACT

BACKGROUND: In recent years, health care has undergone a rapid and unprecedented digital transformation. In many fields of specialty care, such as rheumatology, this shift is driven by the growing number of patients and limited resources, leading to increased use of digital health technologies (DHTs) to maintain high-quality clinical care. Previous studies examined user acceptance of individual DHTs in rheumatology, such as telemedicine, video consultations, and mHealth. However, it is essential to conduct cross-technology and continuous analyses of user acceptance and DHT use to maximize the benefits for all relevant stakeholders.

OBJECTIVE: This study aimed to explore the current acceptance, use, and preferences regarding DHTs among patients in rheumatology care in Germany.

METHODS: Rheumatology patients from 3 clinics in Germany were surveyed to understand their perspectives on DHTs. The survey included main themes, including acceptance, preferences, COVID-19’s impact, potential, and barriers related to DHTs. The data were analyzed using descriptive statistics and correlation analysis.

RESULTS: Out of 337 participants, 53% (179/337) reported using DHTs. Specific technologies included wearables (72/337, 21%), mHealth apps (71/337, 21%), digital therapeutics (32/337, 9%), electronic prescriptions (30/337, 9%), video consultations (15/337, 4%), and at-home blood self-sampling (3/337, 1%). Nearly two-thirds (220/337, 65%) found DHTs useful, and 69% (233/337) held a generally positive attitude toward DHTs. Attitudes shifted positively during the COVID-19 pandemic for 40% (135/337) of participants. Higher education was more prevalent among DHT users (114/179, 63.7%) compared with nonusers (42/151, 27.8%; P=.02). The main potential benefits identified were location-independent use (244/337, 72%) and time-independent use (216/337, 64%). Key barriers included insufficient user knowledge (165/337, 49%) and limited information on DHTs (134/337, 40%).

CONCLUSIONS: Patient acceptance and use of DHTs in rheumatology is increasing in Germany. A prospective, standardized monitoring of digital transformation in rheumatology care is highly needed.

PMID:39761546 | DOI:10.2196/52601

Retina. 2025 Jan 2. doi: 10.1097/IAE.0000000000004394. Online ahead of print.

ABSTRACT

PURPOSE: To report one-year real-world evidence on intraocular inflammation (IOI) adverse events (AEs) in patients undergoing faricimab therapy in a tertiary care hospital.

METHODS: A retrospective review of electronic medical records was conducted for patients receiving faricimab treatment for neovascular age-related macular degeneration (nAMD) and diabetic macular edema (DME) at Moorfields Eye Hospital between September 1st, 2022, and August 31st, 2023. The primary outcome was the incidence of IOI (excluding endophthalmitis).

RESULTS: 2 318 eyes from 1 860 patients were included and underwent a total of 10 297 injections. A total of 20 eyes (16 patients) had ≥ 1 adverse event of IOI. Estimated incidence of IOI was 0.19% per injection (95%CI 0.12-0.30), 0.86% per eye (95% CI 0.53- 1.33] and 0.86% per patient (95%CI 0.49- 1.39). IOI mostly occurred within the first injections (median 3.5 injections, range 1-10). All cases presented with anterior uveitis and were associated with vitritis in 4 eyes (20%). No cases of posterior uveitis or evidence of retinal vascular occlusion were reported. There was no statistically significant difference between mean visual acuity before and after IOI event (0.40 logMAR and 0.378 logMAR respectively, p = .26).

CONCLUSION: In this real-world report, faricimab was well tolerated with an incidence of IOI-related AEs consistent to that observed in registration trials. The AEs were generally mild and had a favourable prognosis.

PMID:39761510 | DOI:10.1097/IAE.0000000000004394

Eur J Public Health. 2025 Jan 6:ckae220. doi: 10.1093/eurpub/ckae220. Online ahead of print.

ABSTRACT

Maternal mortality remains a critical global health challenge, with 95% of deaths occurring in low-income countries. While progress was made from 2000 to 2015, regions such as Ethiopia continue to experience high maternal mortality rates, impeding the achievement of the sustainable development goal to reduce maternal deaths to 70 per 100 000 live births by 2030. This study evaluated the effectiveness of a Social and Behavior Change Communication (SBCC) intervention to improve maternal health behaviors. A community-randomized trial was conducted in three districts of Jimma Zone, rural Ethiopia, involving 5057 women. Sixteen primary healthcare units were randomly assigned to either the intervention (SBCC) or control (standard care) group. Data on socio-demographics, antenatal care (ANC) visits, maternal health knowledge, attitudes, and institutional childbirth rates were collected at baseline and endline. Statistical analyses included t-tests, effect sizes, and generalized estimating equations. The intervention group demonstrated significant improvements. Maternal health knowledge increased from 5.68 to 7.70 (P < .001, effect size = 0.34), attitudes improved from 37.49 to 39.73 (P < .001, effect size = 0.29), and ANC visits rose from 3.27 to 4.21 (P < .001, effect size = 0.50). Institutional childbirth rates increased from 0.52 to 0.71 (P < .001, effect size = 0.18). ANC attendance (B = 0.082, P = .002) and positive attitudes (B = 0.055, P < .001) were significant predictors of institutional childbirth. The SBCC intervention significantly enhanced maternal health knowledge, attitudes, ANC utilization, and institutional childbirth rates, highlighting the value of community-based strategies in improving maternal health behaviors.

PMID:39761508 | DOI:10.1093/eurpub/ckae220

J Comput Assist Tomogr. 2024 Nov 13. doi: 10.1097/RCT.0000000000001693. Online ahead of print.

ABSTRACT

OBJECTIVES: The aims of the study are to predict lung function impairment in patients with connective tissue disease (CTD)-associated interstitial lung disease (ILD) through computed tomography (CT) quantitative analysis parameters based on CT deep learning model and density threshold method and to assess the severity of the disease in patients with CTD-ILD.

METHODS: We retrospectively collected chest high-resolution CT images and pulmonary function test results from 105 patients with CTD-ILD between January 2021 and December 2023 (patients staged according to the gender-age-physiology [GAP] system), including 46 males and 59 females, with a median age of 64 years. Additionally, we selected 80 healthy controls (HCs) with matched sex and age, who showed no abnormalities in their chest high-resolution CT. Based on our previously developed RDNet analysis model, the proportion of the lung occupied by reticulation, honeycombing, and total interstitial abnormalities in CTD-ILD patients (ILD% = total interstitial abnormal volume/total lung volume) were calculated. Using the Pulmo-3D software with a threshold segmentation method of -260 to -600, the overall interstitial abnormal proportion (AA%) and mean lung density were obtained. The correlations between CT quantitative analysis parameters and pulmonary function indices were evaluated using Spearman or Pearson correlation coefficients. Stepwise multiple linear regression analysis was used to identify the best CT quantitative predictors for different pulmonary function parameters. Independent risk factors for GAP staging were determined using multifactorial logistic regression. The area under the ROC curve (AUC) differentiated between the CTD-ILD groups and HCs, as well as among GAP stages. The Kruskal-Wallis test was used to compare the differences in pulmonary function indices and CT quantitative analysis parameters among CTD-ILD groups.

RESULTS: Among 105 CTD-ILD patients (58 in GAP I, 36 in GAP II, and 11 in GAP III), results indicated that AA% distinguished between CTD-ILD patients and HCs with the highest AUC value of 0.974 (95% confidence interval: 0.955-0.993). With a threshold set at 9.7%, a sensitivity of 98.7% and a specificity of 89.5% were observed. Both honeycombing and ILD% showed statistically significant correlations with pulmonary function parameters, with honeycombing displaying the highest correlation coefficient with Composite Physiologic Index (CPI, r = 0.612). Multiple linear regression results indicated honeycombing was the best predictor for both the Dlco% and the CPI. Furthermore, multivariable logistic regression analysis identified honeycombing as an independent risk factor for GAP staging. Honeycombing differentiated between GAP I and GAP II + III with the highest AUC value of 0.729 (95% confidence interval: 0.634-0.811). With a threshold set at 8.0%, a sensitivity of 79.3% and a specificity of 57.4% were observed. Significant differences in honeycombing and ILD% were also noted among the disease groups (P < 0.05).

CONCLUSIONS: An AA% of 9.7% was the optimal threshold for differentiating CTD-ILD patients from HCs. Honeycombing can preliminarily predict lung function impairment and was an independent risk factor for GAP staging, offering significant clinical guidance for assessing the severity of the patient’s disease.

PMID:39761506 | DOI:10.1097/RCT.0000000000001693

J Clin Oncol. 2025 Jan 6:JCO2401710. doi: 10.1200/JCO-24-01710. Online ahead of print.

ABSTRACT

PURPOSE: ERBB2 overexpression/amplification in RAS/BRAF wild-type (WT) metastatic colorectal cancer (mCRC; human epidermal growth factor receptor 2 [HER2]-positive mCRC) appears to be associated with limited benefit from anti-EGFR antibodies and promising responses to dual-HER2 inhibition; however, comparative efficacy has not been investigated. We conducted a randomized phase II trial to evaluate efficacy and safety of dual-HER2 inhibition against standard-of-care anti-EGFR antibody-based therapy as second/third-line treatment in HER2-positive mCRC.

METHODS: Patients with RAS/BRAF-WT mCRC after central confirmation of HER2 positivity (immunohistochemistry 3+ or 2+ and in situ hybridization amplified [HER2/CEP17 ratio >2.0]) were assigned (1:1) to either trastuzumab plus pertuzumab (TP; trastuzumab 6 mg/kg and pertuzumab 420 mg once every 3 weeks) or cetuximab plus irinotecan (CETIRI; cetuximab 500 mg/m² and irinotecan 180 mg/m² once every 2 weeks) until progression or unacceptable toxicity. Crossover to TP was allowed after progression on CETIRI. The primary end point was progression-free survival (PFS). Secondary end points included objective response rate (ORR), overall survival, safety, and HER2 gene copy number (GCN ≥20/<20) as a predictive factor.

RESULTS: Between October 2017 and March 2022, 54 participants were assigned to TP (n = 26) and CETIRI (n = 28). Median PFS did not vary significantly by treatment: 4.7 (95% CI, 1.9 to 7.6) and 3.7 (95% CI, 1.6 to 6.7) months in the TP and CETIRI groups, respectively. Efficacy of TP versus CETIRI differed significantly by HER2 GCN (median PFS, GCN ≥20 [9.9 v 2.9 months] and GCN <20 [3.0 v 4.2 months], respectively; P interaction = .003). On TP, ORR was 34.6% (57.1% with GCN ≥20 v 9.1% with GCN <20) with median GCN of 29.7 versus 13.2 for responders and nonresponders, respectively (P = .004). Grade ≥3 adverse events occurred in 23.1% and 46.1% of participants with TP and CETIRI, respectively.

CONCLUSION: TP appears to be a safe and effective cytotoxic chemotherapy-free option for patients with RAS/BRAF-WT, HER2-positive mCRC. Higher levels of HER2 amplification were associated with greater degree of clinical benefit from TP vis-à-vis CETIRI.

PMID:39761503 | DOI:10.1200/JCO-24-01710

J Comput Assist Tomogr. 2024 Nov 13. doi: 10.1097/RCT.0000000000001682. Online ahead of print.

ABSTRACT

OBJECTIVE: Computed tomography (CT) measured muscle density is prognostic of health outcomes. However, the use of intravenous contrast obscures prognoses by artificially increasing CT muscle density. We previously established a correction to equalize contrast and noncontrast muscle density measurements. While this correction was validated internally, the objective of this study was to obtain external validation using different patient cohorts, muscle regions, and CT series.

METHODS: CT images from 109 patients with kidney tumors who received abdominal CT scans with a multiphase intravenous contrast protocol were analyzed. Paraspinal muscle density measurements taken during noncontrast, venous phase, and delayed phase contrast scans were collected. An a priori correction of -7.5 Hounsfield units (HU) was applied to muscle measurements. Equivalence testing was utilized to determine statistical similarity.

RESULTS: In the sample of 109 patients (mean age: 63 years [SD: 14.3]; 41.3% female), densities in smaller regions of interest within the paraspinal muscles and the entire paraspinal muscle density (PS) in venous and delayed phase contrast scans were higher than in noncontrast. Equivalence testing showed that average corrected contrast and noncontrast muscle densities were within 3 HU for both muscle measures for the total patient sample, and for a majority of male and female subsamples. The correction is suitable for regions of interests of venous contrast (90% CI: -1.90, -0.69 HU) and delayed contrast scans (90% CI: 0.075, 1.29 HU) and within the PS measures of venous contrast (90% CI: -2.04, -0.94 HU) and delayed contrast scans (90% CI: -0.11, 0.89 HU).

CONCLUSIONS: The previously established correction for contrast of -7.5 HU was applied in a new patient population, axial muscle region, muscle measurement size, and expanded on previously studied contrast phases. The correction produced contrast-corrected muscle densities that were statistically equivalent to noncontrast muscle densities. The simplicity of the correction gives clinicians a tool that seamlessly integrates into practice or research to improve harmonization of data between contrast and noncontrast scans.

PMID:39761492 | DOI:10.1097/RCT.0000000000001682

J Comput Assist Tomogr. 2024 Nov 13. doi: 10.1097/RCT.0000000000001685. Online ahead of print.

ABSTRACT

OBJECTIVE: Discriminating between hemorrhage and iodine extravasation can pose challenges in conventional computed tomography (CCT) images following preoperative embolization for meningioma. This study aimed to assess the efficacy of dual-energy computed tomography (DECT) in differentiating hemorrhage from iodine extravasation after preoperative embolization for meningioma.

METHODS: Twenty-one consecutive meningioma patients who underwent CCT before and DECT immediately after preoperative embolization were included in this study. Two independent observers conducted qualitative assessments on CCT and virtual noncontrast (VNC) images and iodine maps (IMs) to differentiate between hemorrhage and iodine extravasation. One observer recorded CT values of hemorrhage and iodine extravasation on CCT and VNC images. The ratio of maximum attenuation to minimum attenuation on VNC images was defined as the VNC ratio. Statistical analysis included Kappa (κ) statistics, unpaired t tests, and receiver operating characteristic (ROC) analysis.

RESULTS: Interobserver agreement for qualitative assessment was fair (κ = 0.231) for CCT alone and good (κ = 0.723) for CCT plus VNC imaging and IM. The addition of VNC imaging and IM to CCT improved differential confidence in 16 (76%) and 18 (86%) cases of the two observers, respectively, increasing the area under the receiver operating characteristic curve (AUROC) from 0.868 to 0.895 and 0.658 to 0.947, respectively. At a cutoff value of 1.527, the VNC ratio was significantly higher for hemorrhage than iodine extravasation (P < 0.05), with the highest diagnostic performance (AUROC, 1).

CONCLUSIONS: DECT with VNC imaging and IM is useful for differentiating hemorrhage from iodine extravasation in meningiomas with preoperative embolization.

PMID:39761489 | DOI:10.1097/RCT.0000000000001685

J Comput Assist Tomogr. 2024 Nov 13. doi: 10.1097/RCT.0000000000001642. Online ahead of print.

ABSTRACT

OBJECTIVE: The aim of the study is to assess the diagnostic performance of quantitative analysis of diffusion-weighted imaging in assessing treatment response in cervical cancer patients.

METHODS: A retrospective analysis was done for 50 patients with locally advanced cervical cancer who received concurrent chemoradiotherapy and underwent magnetic resonance imaging and diffusion-weighted imaging. Treatment response was classified into 4 categories according to RECIST criteria 6 months after therapy completion. Apparent diffusion coefficient (ADC) values were measured using both region of interest (ROI) ADC and whole lesion (WL) ADC histogram for all cases at both baseline pretreatment and posttreatment Magnetic resonance imaging studies. Changes in ADC values were calculated and compared between groups.

RESULTS: The percentage change of ROI-ADCmean at a cutoff value of >20 had excellent discrimination of responders versus nonresponders, while the percentage change of WL-ADCmean, ADCmin, and ADCmax at cutoff values of >12.5, >35.8, and > 19.6 had acceptable discrimination of responders versus nonresponders. Logistic regression analysis revealed that only baseline WL ADCmin was a statistically significant independent predictor of response. Cancer cervix patients with baseline ADCmin < or equal to 0.73 have 12.1 times higher odds of exhibiting a response.

CONCLUSIONS: The percentage change of ROI-ADCmean and WL histogram ADCmean values after concurrent chemoradiotherapy can predict response. Pretreatment WL histogram ADCmin was a statistically significant independent predictor of posttherapy response.

PMID:39761488 | DOI:10.1097/RCT.0000000000001642

J Comput Assist Tomogr. 2024 Nov 18. doi: 10.1097/RCT.0000000000001694. Online ahead of print.

ABSTRACT

BACKGROUND: The latest generation of computed tomography (CT) systems based on photon-counting detector promises significant improvements in several clinical applications, including chest imaging.

PURPOSE: The aim of the study is to evaluate the image quality of ultra-high-resolution (UHR) photon-counting detector CT (PCD-CT) of the lung using four sharp reconstruction kernels.

MATERIAL AND METHODS: This retrospective study included 25 patients (11 women and 14 men; median age, 71 years) who underwent unenhanced chest CT from April to May 2023. Images were acquired in UHR mode on a clinical dual-source PCD-CT scanner and reconstructed with four sharp kernels (Bl64, Br76, Br84, Br96). Quantitative image analysis included the measurement of image noise, and the calculation of signal-to-noise ratio, and contrast-to-noise ratio. Two radiologists independently rated the images on a 5-point Likert scale for image sharpness, image noise, overall image quality, and airway details. The 4 image sets were compared pairwise in the statistical analysis.

RESULTS: Image noise was lowest for Br76 (74.16 ± 22.05, P < 0.001). Signal-to-noise ratio was significantly higher in the Br76 images (13.34 ± 3.47), than in the other 3 image sets (all P < 0.001). The Br76 images demonstrated the highest contrast-to-noise ratio among all reconstructions (1.54 ± 0.86, all P < 0.001). Subjective image sharpness, image noise, overall image quality, and airway detail were best in the Br76 images (all P < 0.001 to P < 0.01, for both readers).

CONCLUSIONS: The use of the Br76 reconstruction kernel provided the best quantitative and qualitative image quality for UHR PCD-CT of the lungs.

PMID:39761487 | DOI:10.1097/RCT.0000000000001694