Sign in to use this feature.

Years

Between: -

Subjects

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Journals

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Article Types

Countries / Regions

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Search Results (2,641)

Search Parameters:
Keywords = area under the ROC curve

Order results
Result details
Results per page
Select all
Export citation of selected articles as:
13 pages, 769 KiB  
Article
A Novel You Only Listen Once (YOLO) Deep Learning Model for Automatic Prominent Bowel Sounds Detection: Feasibility Study in Healthy Subjects
by Rohan Kalahasty, Gayathri Yerrapragada, Jieun Lee, Keerthy Gopalakrishnan, Avneet Kaur, Pratyusha Muddaloor, Divyanshi Sood, Charmy Parikh, Jay Gohri, Gianeshwaree Alias Rachna Panjwani, Naghmeh Asadimanesh, Rabiah Aslam Ansari, Swetha Rapolu, Poonguzhali Elangovan, Shiva Sankari Karuppiah, Vijaya M. Dasari, Scott A. Helgeson, Venkata S. Akshintala and Shivaram P. Arunachalam
Sensors 2025, 25(15), 4735; https://doi.org/10.3390/s25154735 (registering DOI) - 31 Jul 2025
Abstract
Accurate diagnosis of gastrointestinal (GI) diseases typically requires invasive procedures or imaging studies that pose the risk of various post-procedural complications or involve radiation exposure. Bowel sounds (BSs), though typically described during a GI-focused physical exam, are highly inaccurate and variable, with low [...] Read more.
Accurate diagnosis of gastrointestinal (GI) diseases typically requires invasive procedures or imaging studies that pose the risk of various post-procedural complications or involve radiation exposure. Bowel sounds (BSs), though typically described during a GI-focused physical exam, are highly inaccurate and variable, with low clinical value in diagnosis. Interpretation of the acoustic characteristics of BSs, i.e., using a phonoenterogram (PEG), may aid in diagnosing various GI conditions non-invasively. Use of artificial intelligence (AI) and improvements in computational analysis can enhance the use of PEGs in different GI diseases and lead to a non-invasive, cost-effective diagnostic modality that has not been explored before. The purpose of this work was to develop an automated AI model, You Only Listen Once (YOLO), to detect prominent bowel sounds that can enable real-time analysis for future GI disease detection and diagnosis. A total of 110 2-minute PEGs sampled at 44.1 kHz were recorded using the Eko DUO® stethoscope from eight healthy volunteers at two locations, namely, left upper quadrant (LUQ) and right lower quadrant (RLQ) after IRB approval. The datasets were annotated by trained physicians, categorizing BSs as prominent or obscure using version 1.7 of Label Studio Software®. Each BS recording was split up into 375 ms segments with 200 ms overlap for real-time BS detection. Each segment was binned based on whether it contained a prominent BS, resulting in a dataset of 36,149 non-prominent segments and 6435 prominent segments. Our dataset was divided into training, validation, and test sets (60/20/20% split). A 1D-CNN augmented transformer was trained to classify these segments via the input of Mel-frequency cepstral coefficients. The developed AI model achieved area under the receiver operating curve (ROC) of 0.92, accuracy of 86.6%, precision of 86.85%, and recall of 86.08%. This shows that the 1D-CNN augmented transformer with Mel-frequency cepstral coefficients achieved creditable performance metrics, signifying the YOLO model’s capability to classify prominent bowel sounds that can be further analyzed for various GI diseases. This proof-of-concept study in healthy volunteers demonstrates that automated BS detection can pave the way for developing more intuitive and efficient AI-PEG devices that can be trained and utilized to diagnose various GI conditions. To ensure the robustness and generalizability of these findings, further investigations encompassing a broader cohort, inclusive of both healthy and disease states are needed. Full article
(This article belongs to the Special Issue Biomedical Signals, Images and Healthcare Data Analysis: 2nd Edition)
Show Figures

Figure 1

7 pages, 266 KiB  
Communication
Respiratory Failure in Patients with Intracerebral Hemorrhage and Intraventricular Hemorrhage Extension: A Retrospective Study
by Min Cheol Chang, Michael Y. Lee, Sang Gyu Kwak and Ah Young Lee
Healthcare 2025, 13(15), 1876; https://doi.org/10.3390/healthcare13151876 (registering DOI) - 31 Jul 2025
Abstract
Background/Objectives: This study aimed to identify the risk factors for respiratory failure in patients with intracerebral hemorrhage (ICH) accompanied by intraventricular hemorrhage (IVH) extension. Methods: We retrospectively included 208 patients with ICH accompanied by IVH extension. Respiratory failure was defined as carbon [...] Read more.
Background/Objectives: This study aimed to identify the risk factors for respiratory failure in patients with intracerebral hemorrhage (ICH) accompanied by intraventricular hemorrhage (IVH) extension. Methods: We retrospectively included 208 patients with ICH accompanied by IVH extension. Respiratory failure was defined as carbon dioxide levels > 45 mmHg with a pH < 7.35 in arterial blood gas analysis (ABGA) or the application of a ventilator due to respiratory dysfunction. We measured the severity of IVH extension using the Graeb scale, and ICH volume was assessed for each patient. Results: Of the 208 included patients, 83 had respiratory failure. There were no significant differences in age, sex ratio, or Graeb scale score between patients with and without respiratory failure (p > 0.05). However, ICH volume was significantly larger in patients with respiratory failure (42.0 ± 42.5 mL) than in those without (26.4 ± 25.7 mL) (p = 0.003). In the receiver operating characteristic (ROC) curve analysis, the area under the ROC curve for ICH volume predicting respiratory failure was 0.612. The optimal threshold for detecting respiration failure in patients with ICH and IVH dilatation, based on the Youden index, was >63.2 mL, with a sensitivity of 30.12% and a specificity of 89.60%. Approximately 40% of patients experienced respiratory failure following ICH accompanied by IVH extension. Conclusions: A large ICH volume was associated with the occurrence of respiratory failure. Therefore, caution is required in patients with an ICH volume > 63.2 mL. Full article
(This article belongs to the Section Community Care)
Show Figures

Figure 1

12 pages, 1464 KiB  
Article
Improving Prognostic Accuracy of MASCC Score with Lactate and CRP Measurements in Febrile Neutropenic Patients
by Efe Kanter, Ecem Ermete Güler, Süleyman Kırık, Tutku Duman Şahan, Melisa Buse Baygın, Emine Altınöz, Ejder Saylav Bora and Zeynep Karakaya
Diagnostics 2025, 15(15), 1922; https://doi.org/10.3390/diagnostics15151922 (registering DOI) - 31 Jul 2025
Abstract
Objectives: Febrile neutropenia is a common oncologic emergency with significant morbidity and mortality. Although the MASCC (Multinational Association for Supportive Care in Cancer) score is widely used for risk stratification, its limited sensitivity and lack of laboratory parameters reduce its prognostic utility. [...] Read more.
Objectives: Febrile neutropenia is a common oncologic emergency with significant morbidity and mortality. Although the MASCC (Multinational Association for Supportive Care in Cancer) score is widely used for risk stratification, its limited sensitivity and lack of laboratory parameters reduce its prognostic utility. This study aimed to evaluate whether incorporating serum lactate and CRP measurements into the MASCC score enhances its predictive performance for hospital admission and the 30-day mortality. Methods: This retrospective diagnostic accuracy study included adult patients diagnosed with febrile neutropenia in the emergency department of a tertiary care hospital between January 2021 and December 2024. The original MASCC score was calculated, and three modified models were derived: the MASCC-L (lactate/MASCC), MASCC-C (CRP/MASCC) and MASCC-LC models (CRP × lactate/MASCC). The predictive accuracy for hospital admission and the 30-day all-cause mortality was assessed using ROC analysis. Results: A total of 269 patients (mean age: 67.6 ± 12.4 years) were included; the 30-day mortality was 3.0%. The MASCC-LC model demonstrated the highest discriminative ability for mortality prediction (area under the curve (AUC): 0.995; sensitivity: 100%; specificity: 98%). For hospital admission prediction, the MASCC-C model had the highest specificity (81%), while the MASCC-LC model showed the best balance of sensitivity and specificity (both 73%). All the modified models outperformed the original MASCC score regarding both endpoints. Conclusions: Integrating lactate and CRP measurements into the MASCC score significantly improves its prognostic accuracy for both mortality and hospital admission in febrile neutropenic patients. The MASCC-LC model, relying on only three objective parameters, may serve as a practical and efficient tool for early risk stratification in emergency settings. Full article
(This article belongs to the Special Issue Clinical Diagnosis and Management in Emergency and Hospital Medicine)
Show Figures

Figure 1

34 pages, 17155 KiB  
Article
Machine Learning Ensemble Methods for Co-Seismic Landslide Susceptibility: Insights from the 2015 Nepal Earthquake
by Tulasi Ram Bhattarai and Netra Prakash Bhandary
Appl. Sci. 2025, 15(15), 8477; https://doi.org/10.3390/app15158477 (registering DOI) - 30 Jul 2025
Abstract
The Mw 7.8 Gorkha Earthquake of 25 April 2015 triggered over 25,000 landslides across central Nepal, with 4775 events concentrated in Gorkha District alone. Despite substantial advances in landslide susceptibility mapping, existing studies often overlook the compound role of post-seismic rainfall and lack [...] Read more.
The Mw 7.8 Gorkha Earthquake of 25 April 2015 triggered over 25,000 landslides across central Nepal, with 4775 events concentrated in Gorkha District alone. Despite substantial advances in landslide susceptibility mapping, existing studies often overlook the compound role of post-seismic rainfall and lack robust spatial validation. To address this gap, we validated an ensemble machine learning framework for co-seismic landslide susceptibility modeling by integrating seismic, geomorphological, hydrological, and anthropogenic variables, including cumulative post-seismic rainfall. Using a balanced dataset of 4775 landslide and non-landslide instances, we evaluated the performance of Logistic Regression (LR), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost) models through spatial cross-validation, SHapley Additive exPlanations (SHAP) explainability, and ablation analysis. The RF model outperformed all others, achieving an accuracy of 87.9% and a Receiver Operating Characteristic (ROC) Area Under the Curve (AUC) value of 0.94, while XGBoost closely followed (AUC = 0.93). Ensemble models collectively classified over 95% of observed landslides into High and Very High susceptibility zones, demonstrating strong spatial reliability. SHAP analysis identified elevation, proximity to fault, peak ground acceleration (PGA), slope, and rainfall as dominant predictors. Notably, the inclusion of post-seismic rainfall substantially improved recall and F1 scores in ablation experiments. Spatial cross-validation revealed the superior generalizability of ensemble models under heterogeneous terrain conditions. The findings underscore the value of integrating post-seismic hydrometeorological factors and spatial validation into susceptibility assessments. We recommend adopting ensemble models, particularly RF, for operational hazard mapping in earthquake-prone mountainous regions. Future research should explore the integration of dynamic rainfall thresholds and physics-informed frameworks to enhance early warning systems and climate resilience. Full article
(This article belongs to the Section Earth Sciences)
Show Figures

Figure 1

16 pages, 3482 KiB  
Article
Reliability of Automated Amyloid PET Quantification: Real-World Validation of Commercial Tools Against Centiloid Project Method
by Yeon-koo Kang, Jae Won Min, Soo Jin Kwon and Seunggyun Ha
Tomography 2025, 11(8), 86; https://doi.org/10.3390/tomography11080086 - 30 Jul 2025
Abstract
Background: Despite the growing demand for amyloid PET quantification, practical challenges remain. As automated software platforms are increasingly adopted to address these limitations, we evaluated the reliability of commercial tools for Centiloid quantification against the original Centiloid Project method. Methods: This retrospective study [...] Read more.
Background: Despite the growing demand for amyloid PET quantification, practical challenges remain. As automated software platforms are increasingly adopted to address these limitations, we evaluated the reliability of commercial tools for Centiloid quantification against the original Centiloid Project method. Methods: This retrospective study included 332 amyloid PET scans (165 [18F]Florbetaben; 167 [18F]Flutemetamol) performed for suspected mild cognitive impairments or dementia, paired with T1-weighted MRI within one year. Centiloid values were calculated using three automated software platforms, BTXBrain, MIMneuro, and SCALE PET, and compared with the original Centiloid method. The agreement was assessed using Pearson’s correlation coefficient, the intraclass correlation coefficient (ICC), a Passing–Bablok regression, and Bland–Altman plots. The concordance with the visual interpretation was evaluated using receiver operating characteristic (ROC) curves. Results: BTXBrain (R = 0.993; ICC = 0.986) and SCALE PET (R = 0.992; ICC = 0.991) demonstrated an excellent correlation with the reference, while MIMneuro showed a slightly lower agreement (R = 0.974; ICC = 0.966). BTXBrain exhibited a proportional underestimation (slope = 0.872 [0.860–0.885]), MIMneuro showed a significant overestimation (slope = 1.053 [1.026–1.081]), and SCALE PET demonstrated a minimal bias (slope = 1.014 [0.999–1.029]). The bias pattern was particularly noted for FMM. All platforms maintained their trends for correlations and biases when focusing on subthreshold-to-low-positive ranges (0–50 Centiloid units). However, all platforms showed an excellent agreement with the visual interpretation (areas under ROC curves > 0.996 for all). Conclusions: Three automated platforms demonstrated an acceptable reliability for Centiloid quantification, although software-specific biases were observed. These differences did not impair their feasibility in aiding the image interpretation, as supported by the concordance with visual readings. Nevertheless, users should recognize the platform-specific characteristics when applying diagnostic thresholds or interpreting longitudinal changes. Full article
(This article belongs to the Section Brain Imaging)
Show Figures

Figure 1

16 pages, 1182 KiB  
Article
Machine Learning-Based Identification of Risk Factors for ICU Mortality in 8902 Critically Ill Patients with Pandemic Viral Infection
by Elisabeth Papiol, Ricard Ferrer, Juan C. Ruiz-Rodríguez, Emili Díaz, Rafael Zaragoza, Marcio Borges-Sa, Julen Berrueta, Josep Gómez, María Bodí, Susana Sancho, Borja Suberviola, Sandra Trefler and Alejandro Rodríguez
J. Clin. Med. 2025, 14(15), 5383; https://doi.org/10.3390/jcm14155383 - 30 Jul 2025
Abstract
Background/Objectives: The SARS-CoV-2 and influenza A (H1N1)pdm09 pandemics have resulted in high numbers of ICU admissions, with high mortality. Identifying risk factors for ICU mortality at the time of admission can help optimize clinical decision making. However, the risk factors identified may [...] Read more.
Background/Objectives: The SARS-CoV-2 and influenza A (H1N1)pdm09 pandemics have resulted in high numbers of ICU admissions, with high mortality. Identifying risk factors for ICU mortality at the time of admission can help optimize clinical decision making. However, the risk factors identified may differ, depending on the type of analysis used. Our aim is to compare the risk factors and performance of a linear model (multivariable logistic regression, GLM) with a non-linear model (random forest, RF) in a large national cohort. Methods: A retrospective analysis was performed on a multicenter database including 8902 critically ill patients with influenza A (H1N1)pdm09 or COVID-19 admitted to 184 Spanish ICUs. Demographic, clinical, laboratory, and microbiological data from the first 24 h were used. Prediction models were built using GLM and RF. The performance of the GLM was evaluated by area under the ROC curve (AUC), precision, sensitivity, and specificity, while the RF by out-of-bag (OOB) error and accuracy. In addition, in the RF, the im-portance of the variables in terms of accuracy reduction (AR) and Gini index reduction (GI) was determined. Results: Overall mortality in the ICU was 25.8%. Model performance was similar, with AUC = 76% for GLM, and AUC = 75.6% for RF. GLM identified 17 independent risk factors, while RF identified 19 for AR and 23 for GI. Thirteen variables were found to be important in both models. Laboratory variables such as procalcitonin, white blood cells, lactate, or D-dimer levels were not significant in GLM but were significant in RF. On the contrary, acute kidney injury and the presence of Acinetobacter spp. were important variables in the GLM but not in the RF. Conclusions: Although the performance of linear and non-linear models was similar, different risk factors were determined, depending on the model used. This alerts clinicians to the limitations and usefulness of studies limited to a single type of model. Full article
(This article belongs to the Special Issue Current Trends and Prospects of Critical Emergency Medicine)
Show Figures

Figure 1

11 pages, 349 KiB  
Article
Sepsis Prediction: Biomarkers Combined in a Bayesian Approach
by João V. B. Cabral, Maria M. B. M. da Silveira, Wilma T. F. Vasconcelos, Amanda T. Xavier, Fábio H. P. C. de Oliveira, Thaysa M. G. A. L. de Menezes, Keylla T. F. Barbosa, Thaisa R. Figueiredo, Jabiael C. da Silva Filho, Tamara Silva, Leuridan C. Torres, Dário C. Sobral Filho and Dinaldo C. de Oliveira
Int. J. Mol. Sci. 2025, 26(15), 7379; https://doi.org/10.3390/ijms26157379 - 30 Jul 2025
Abstract
Sepsis is a serious public health problem. sTREM-1 is a marker of inflammatory and infectious processes that has the potential to become a useful tool for predicting the evolution of sepsis. A prediction model for sepsis was constructed by combining sTREM-1, CRP, and [...] Read more.
Sepsis is a serious public health problem. sTREM-1 is a marker of inflammatory and infectious processes that has the potential to become a useful tool for predicting the evolution of sepsis. A prediction model for sepsis was constructed by combining sTREM-1, CRP, and a leukogram via a Bayesian network. A translational study carried out with 32 children with congenital heart disease who had undergone surgical correction at a public referral hospital in Northeast Brazil. In the postoperative period, the mean value of sTREM-1 was greater among patients diagnosed with sepsis than among those not diagnosed with sepsis (394.58 pg/mL versus 239.93 pg/mL, p < 0.001). Analysis of the ROC curve for sTREM-1 and sepsis revealed that the area under the curve was 0.761, with a 95% CI (0.587–0.935) and p = 0.013. With the Bayesian model, we found that a 100% probability of sepsis was related to postoperative blood concentrations of CRP above 71 mg/dL, a leukogram above 14,000 cells/μL, and sTREM-1 concentrations above the cutoff point (283.53 pg/mL). The proposed model using the Bayesian network approach with the combination of CRP, leukocyte count, and postoperative sTREM-1 showed promise for the diagnosis of sepsis. Full article
Show Figures

Figure 1

10 pages, 714 KiB  
Article
Use of Mid-Upper Arm Circumference Band in Wasting Detection in Children with Cerebral Palsy in Türkiye
by Uğur Topçu, Çiğdem Lazoğlu, Caner Aslan, Abdurrahman Zarif Güney, Zübeyr Kavcar and Orhan Coşkun
Children 2025, 12(8), 1002; https://doi.org/10.3390/children12081002 - 30 Jul 2025
Viewed by 55
Abstract
Background/Objectives: Malnutrition is a common problem in children with cerebral palsy (CP). The aim of this study was to investigate the suitability and diagnostic performance of mid-upper arm circumference (MUAC) z-score in diagnosing wasting in children with CP, and its impact on [...] Read more.
Background/Objectives: Malnutrition is a common problem in children with cerebral palsy (CP). The aim of this study was to investigate the suitability and diagnostic performance of mid-upper arm circumference (MUAC) z-score in diagnosing wasting in children with CP, and its impact on diagnostic accuracy when evaluated concomitantly with additional clinical factors (birth weight, history of phototherapy). Methods: This single-center, cross-sectional study included 83 children with CP, aged 6 months–17 years, followed-up in our clinic. Anthropometric measurements (MUAC, Body Mass Index (BMI)) and clinical data (birth weight, history of phototherapy, Gross Motor Function Classification System (GMFCS)) were prospectively collected. Wasting was defined according to the BMI z-score ≤ −2 criteria. The diagnostic performance of MUAC z-score was evaluated by Receiver Operating Characteristic (ROC) analysis. The contribution of additional covariates was examined using logistic regression analysis and the backward elimination method. Results: MUAC z-score alone demonstrated good discrimination in diagnosing wasting with an Area Under the Curve (AUC) value between 0.805 and 0.821, but its sensitivity was limited (67.0%). No statistically significant difference was found in diagnostic performance between MUAC measurements of the right arm, left arm, and the unaffected arm (p > 0.050). In logistic regression analysis, MUAC z-score (p = 0.001), birth weight (p = 0.014), and a history of phototherapy (p = 0.046) were found to be significantly associated with wasting malnutrition. The simplified model including these variables yielded an AUC value of 0.876. Conclusions: MUAC z-score is a usable tool for wasting malnutrition screening in children with CP. Although its sensitivity is limited when used alone, its diagnostic accuracy increases when evaluated concomitantly with additional clinical factors such as birth weight and a history of phototherapy. This combined approach may offer clinicians a more robust tool for the early diagnosis and management of wasting malnutrition in children with CP. Full article
(This article belongs to the Section Pediatric Neurology & Neurodevelopmental Disorders)
Show Figures

Figure 1

14 pages, 724 KiB  
Article
Fibroblast Growth Factor 23 Is a Strong Predictor of Adverse Events After Left Ventricular Assist Device Implantation
by Wissam Yared, Leyla Dogan, Ahsannullah Madad Fassli, Ajay Moza, Andreas Goetzenich, Christian Stoppe, Ahmed F. A. Mohammed, Sandra Kraemer, Lachmandath Tewarie, Ahmad Abugameh and Rachad Zayat
J. Cardiovasc. Dev. Dis. 2025, 12(8), 290; https://doi.org/10.3390/jcdd12080290 - 29 Jul 2025
Viewed by 99
Abstract
Heart failure (HF) and left ventricular hypertrophy (LVH) are linked to fibroblast growth factor 23 (FGF23). This study aims to analyze whether FGF23 can predict postoperative outcomes in unselected left ventricular assist device (LVAD) candidates. Methods: We conducted a prospective observational study that [...] Read more.
Heart failure (HF) and left ventricular hypertrophy (LVH) are linked to fibroblast growth factor 23 (FGF23). This study aims to analyze whether FGF23 can predict postoperative outcomes in unselected left ventricular assist device (LVAD) candidates. Methods: We conducted a prospective observational study that included 27 patients (25 HeartMate3 and 2 HeartMateII) with a median follow-up of 30 months. We measured preoperative FGF23 plasma levels and computed the HeartMateII risk score (HMRS), the HeartMate3 risk score (HM3RS) and the EuroSCOREII with respect to postoperative mortality, as well as the Michigan right heart failure risk score (MRHFS), the Euromacs RHF risk score (EURORHFS), the CRITT score with respect to RHF prediction and the kidney failure risk equation (KFRE) with respect to kidney failure. Multivariate logistic regression and receiver operating characteristic (ROC) analyses were performed. Results: In the multivariate logistic regression, preoperative FGF23 level was found to be a predictor of postoperative RHF (OR: 1.37, 95-CI: 0.78–2.38; p = 0.031), mortality (OR: 1.10, 95%-CI: 0.90–1.60; p = 0.025) and the need for postoperative dialysis (OR: 1.09, 95%-CI: 0.91–1.44; p = 0.032). In the ROC analysis, FGF23 as a predictor of post-LVAD RHF had an area under the curve (AUC) of 0.81. Conclusions: FGF23 improves the prediction of clinically significant patient outcomes—such as need for dialysis, RHF and mortality—after HM3 and HMII implantation, as adding FGF23 to established risk scores increased their predictive value. Full article
(This article belongs to the Section Cardiovascular Clinical Research)
Show Figures

Figure 1

27 pages, 8496 KiB  
Article
Comparative Performance of Machine Learning Models for Landslide Susceptibility Assessment: Impact of Sampling Strategies in Highway Buffer Zone
by Zhenyu Tang, Shumao Qiu, Haoying Xia, Daming Lin and Mingzhou Bai
Appl. Sci. 2025, 15(15), 8416; https://doi.org/10.3390/app15158416 - 29 Jul 2025
Viewed by 104
Abstract
Landslide susceptibility assessment is critical for hazard mitigation and land-use planning. This study evaluates the impact of two different non-landslide sampling methods—random sampling and sampling constrained by the Global Landslide Hazard Map (GLHM)—on the performance of various machine learning and deep learning models, [...] Read more.
Landslide susceptibility assessment is critical for hazard mitigation and land-use planning. This study evaluates the impact of two different non-landslide sampling methods—random sampling and sampling constrained by the Global Landslide Hazard Map (GLHM)—on the performance of various machine learning and deep learning models, including Naïve Bayes (NB), Support Vector Machine (SVM), SVM-Random Forest hybrid (SVM-RF), and XGBoost. The study area is a 2 km buffer zone along the Duku Highway in Xinjiang, China, with 102 landslide and 102 non-landslide points extracted by aforementioned sampling methods. Models were tested using ROC curves and non-parametric significance tests based on 20 repetitions of 5-fold spatial cross-validation data. GLHM sampling consistently improved AUROC and accuracy across all models (e.g., AUROC gains: NB +8.44, SVM +7.11, SVM–RF +3.45, XGBoost +3.04; accuracy gains: NB +11.30%, SVM +8.33%, SVM–RF +7.40%, XGBoost +8.31%). XGBoost delivered the best performance under both sampling strategies, reaching 94.61% AUROC and 84.30% accuracy with GLHM sampling. SHAP analysis showed that GLHM sampling stabilized feature importance rankings, highlighting STI, TWI, and NDVI as the main controlling factors for landslides in the study area. These results highlight the importance of hazard-informed sampling to enhance landslide susceptibility modeling accuracy and interpretability. Full article
Show Figures

Figure 1

18 pages, 1621 KiB  
Article
Inflammatory Metabolic Index and Metabolic-Inflammatory Stress Index as New Biomarkers for Complicated and Perforated Acute Appendicitis
by Sidere M. Zorrilla-Alfaro, Nestor A. Lechuga-Garcia, Arturo Araujo-Conejo, Leticia A. Ramirez-Hernandez, Idalia Garza-Veloz, Alejandro Mauricio-Gonzalez, Ivan Delgado-Enciso, Iram P. Rodriguez-Sanchez and Margarita L. Martinez-Fierro
J. Clin. Med. 2025, 14(15), 5281; https://doi.org/10.3390/jcm14155281 - 25 Jul 2025
Viewed by 409
Abstract
Background: Acute appendicitis is a common emergency requiring abdominal surgery. Despite its prevalence, there are no specific biomarkers for its diagnosis and prognosis. The aim of this study was to assess the basic laboratory tests of patients with acute appendicitis and to [...] Read more.
Background: Acute appendicitis is a common emergency requiring abdominal surgery. Despite its prevalence, there are no specific biomarkers for its diagnosis and prognosis. The aim of this study was to assess the basic laboratory tests of patients with acute appendicitis and to evaluate and integrate biochemical variables into the diagnosis of appendicitis. Methods: This was a retrospective, cross-sectional cohort study that included data from patients who underwent an appendectomy. Two groups of patients were considered based on their surgical (non-complicated/complicated appendicitis) or pathological diagnosis (non-perforated/perforated appendicitis). Factor analysis was carried out to identify communalities to put forward classificatory indices. Receiver operating characteristic (ROC) analysis was used to assess the accuracy of the predictions. Results: The cohort included 246 patients (51.6% male, mean age: 24.79 ± 19.32 years). By using their biochemical data, we generated 6 new indices whose areas under the ROC curve (AUC) ranged between 0.632 and 0.762 for complicated appendicitis and from 0.597 to 0.742 for perforated appendicitis. Inflammatory Metabolic Index (IMI) at the fixed cutoffs was a promising biomarker for both histopathological and surgical diagnoses with odds ratios (OR) of 10.45 and 5.21, respectively. The Metabolic-Inflammatory Stress Index (MISI) showed high specificity (over 72%) and significant AUC values for both diagnoses (0.742 and 0.676). These findings were reinforced by significant p-values and Youden indices. Conclusions: IMI and MISI were demonstrated to be effective biomarkers for complicated and perforated appendicitis. IMI provides predictive capability, while MISI offers specificity and significant AUC values for both histopathological and surgical diagnoses. Incorporating these biomarkers could enhance the accuracy of appendicitis diagnosis and potentially guide clinical decision-making. Full article
(This article belongs to the Section Gastroenterology & Hepatopancreatobiliary Medicine)
Show Figures

Figure 1

14 pages, 12586 KiB  
Article
Three-Dimensional Mineralization Prediction Using the Information Value Method: A Case Study of the Bainiuchang Ag Polymetallic Deposit, Southwest China
by Fuju Jia, Guolong Zheng, Guangzhi Meng, Long Jian, He Chang, Ping Pan, Jianguo Gao, Zhixiao Wu and Ceting Yang
Minerals 2025, 15(8), 783; https://doi.org/10.3390/min15080783 - 25 Jul 2025
Viewed by 192
Abstract
Three-dimensional geological modeling combined with the information value method offers a powerful approach for predicting mineralization in complex geological settings. This study applies these techniques to the Bainiuchang Ag polymetallic deposit in Southwest China. We constructed a detailed 3D geological model to identify [...] Read more.
Three-dimensional geological modeling combined with the information value method offers a powerful approach for predicting mineralization in complex geological settings. This study applies these techniques to the Bainiuchang Ag polymetallic deposit in Southwest China. We constructed a detailed 3D geological model to identify key geological factors influencing mineralization, such as the F3 fault, secondary faults, and the middle Cambrian Tianpeng Formation. Using the information value method, we evaluated the mineralization potential of each unit in the study area. Receiver operating characteristic (ROC) curves and fractal analysis were employed to determine an optimal threshold (information value > 2.6) for delineating high-potential exploration targets. Our results identified three priority targets (A1, A2, and A3), demonstrating the effectiveness of integrating 3D modeling with probabilistic mineral prospectivity mapping. This approach provides a robust framework for deep and peripheral exploration in structurally complex deposits, offering valuable insights for future mineral exploration efforts. Full article
Show Figures

Figure 1

24 pages, 9767 KiB  
Article
Improved Binary Classification of Underwater Images Using a Modified ResNet-18 Model
by Mehrunnisa, Mikolaj Leszczuk, Dawid Juszka and Yi Zhang
Electronics 2025, 14(15), 2954; https://doi.org/10.3390/electronics14152954 - 24 Jul 2025
Viewed by 252
Abstract
In recent years, the classification of underwater images has become one of the most remarkable areas of research in computer vision due to its useful applications in marine sciences, aquatic robotics, and sea exploration. Underwater imaging is pivotal for the evaluation of marine [...] Read more.
In recent years, the classification of underwater images has become one of the most remarkable areas of research in computer vision due to its useful applications in marine sciences, aquatic robotics, and sea exploration. Underwater imaging is pivotal for the evaluation of marine eco-systems, analysis of biological habitats, and monitoring underwater infrastructure. Extracting useful information from underwater images is highly challenging due to factors such as light distortion, scattering, poor contrast, and complex foreground patterns. These difficulties make traditional image processing and machine learning techniques struggle to analyze images accurately. As a result, these challenges and complexities make the classification difficult or poor to perform. Recently, deep learning techniques, especially convolutional neural network (CNN), have emerged as influential tools for underwater image classification, contributing noteworthy improvements in accuracy and performance in the presence of all these challenges. In this paper, we have proposed a modified ResNet-18 model for the binary classification of underwater images into raw and enhanced images. In the proposed modified ResNet-18 model, we have added new layers such as Linear, rectified linear unit (ReLU) and dropout layers, arranged in a block that was repeated three times to enhance feature extraction and improve learning. This enables our model to learn the complex patterns present in the image in more detail, which helps the model to perform the classification very well. Due to these newly added layers, our proposed model addresses various complexities such as noise, distortion, varying illumination conditions, and complex patterns by learning vigorous features from underwater image datasets. To handle the issue of class imbalance present in the dataset, we applied a data augmentation technique. The proposed model achieved outstanding performance, with 96% accuracy, 99% precision, 92% sensitivity, 99% specificity, 95% F1-score, and a 96% Area under the Receiver Operating Characteristic Curve (AUC-ROC) score. These results demonstrate the strength and reliability of our proposed model in handling the challenges posed by the underwater imagery and making it a favorable solution for advancing underwater image classification tasks. Full article
Show Figures

Figure 1

15 pages, 3635 KiB  
Article
Comparison of Apparent Diffusion Coefficient Values on Diffusion-Weighted MRI for Differentiating Hepatocellular Carcinoma and Intrahepatic Cholangiocarcinoma
by Katrīna Marija Konošenoka, Nauris Zdanovskis, Aina Kratovska, Artūrs Šilovs and Veronika Zaiceva
Diagnostics 2025, 15(15), 1861; https://doi.org/10.3390/diagnostics15151861 - 24 Jul 2025
Viewed by 278
Abstract
Background and Objectives: Accurate noninvasive differentiation between hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (ICC) remains a clinical challenge. This study aimed to assess the dignostic performance of apparent diffusion coefficient (ADC) values from diffusion-weighted MRI in distinguishing between HCC and ICC, with [...] Read more.
Background and Objectives: Accurate noninvasive differentiation between hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (ICC) remains a clinical challenge. This study aimed to assess the dignostic performance of apparent diffusion coefficient (ADC) values from diffusion-weighted MRI in distinguishing between HCC and ICC, with histological confirmation as the gold standard. Materials and Methods: A retrospective analysis was performed on 61 patients (41 HCC, 20 ICC) who underwent liver MRI and percutaneous biopsy between 2019 and 2024. ADC values were measured from diffusion-weighted sequences (b-values of 0, 500, and 1000 s/mm2), and regions of interest were placed over solid tumor areas. Statistical analyses included t-tests, one-way ANOVA, and ROC curve analysis. Results: Mean ADC values did not differ significantly between HCC (1.09 ± 0.19 × 10−3 mm2/s) and ICC (1.08 ± 0.11 × 10−3 mm2/s). ROC analysis showed poor discriminative ability (AUC = 0.520; p = 0.806). In HCC, ADC values decreased with lower differentiation grades (p = 0.008, η2 = 0.224). No significant trend was observed in ICC (p = 0.410, η2 = 0.100). Immunohistochemical markers such as CK-7, Glypican 3, and TTF-1 showed significant diagnostic value between tumor subtypes. Conclusions: ADC values have limited utility for distinguishing HCC from ICC but may aid in HCC grading. Immunohistochemistry remains essential for accurate diagnosis, especially in poorly differentiated tumors. Further studies with larger cohorts are recommended to improve noninvasive diagnostic protocols. Full article
(This article belongs to the Special Issue Diagnostic Imaging in Gastrointestinal and Liver Diseases)
Show Figures

Figure 1

16 pages, 818 KiB  
Article
Predictive Value of Frailty, Comorbidity, and Patient-Reported Measures for Hospitalization or Death in Older Outpatients: Quality of Life and Depression as Prognostic Red Flags
by Dimitrios Anagnostou, Nikolaos Theodorakis, Sofia Kalantzi, Aikaterini Spyridaki, Christos Chitas, Vassilis Milionis, Zoi Kollia, Michalitsa Christodoulou, Ioanna Nella, Aggeliki Spathara, Efi Gourzoulidou, Sofia Athinaiou, Gesthimani Triantafylli, Georgia Vamvakou and Maria Nikolaou
Diagnostics 2025, 15(15), 1857; https://doi.org/10.3390/diagnostics15151857 - 23 Jul 2025
Viewed by 221
Abstract
Objectives: To identify clinical, functional, laboratory, and patient-reported parameters associated with medium-term risk of hospitalization or death among older adults attending a multidisciplinary outpatient clinic, and to assess the predictive performance of these measures for individual risk stratification. Methods: In this [...] Read more.
Objectives: To identify clinical, functional, laboratory, and patient-reported parameters associated with medium-term risk of hospitalization or death among older adults attending a multidisciplinary outpatient clinic, and to assess the predictive performance of these measures for individual risk stratification. Methods: In this cohort study, 350 adults aged ≥65 years were assessed at baseline and followed for an average of 8 months. The primary outcome was a composite of hospitalization or all-cause mortality. Parameters assessed included frailty and comorbidity measures, functional parameters, such as gait speed and grip strength, laboratory biomarkers, and patient-reported measures, such as quality of life (QoL, assessed on a Likert scale) and the presence of depressive symptoms. Predictive performance was evaluated using univariable logistic regression and multivariable modeling. Discriminative ability was assessed via area under the ROC curve (AUC), and selected models were internally validated using repeated k-fold cross-validation. Results: Overall, 40 participants (11.4%) experienced hospitalization or death. Traditional clinical risk indicators, including frailty and comorbidity scores, were significantly associated with the outcome. Patient-reported QoL (AUC = 0.74) and Geriatric Depression Scale (GDS) scores (AUC = 0.67) demonstrated useful overall discriminatory ability, with high specificities at optimal cut-offs, suggesting they could act as “red flags” for adverse outcomes. However, the limited sensitivities of individual predictors underscore the need for more comprehensive screening instruments with improved ability to identify at-risk individuals earlier. A multivariable model that incorporated several predictors did not outperform QoL alone (AUC = 0.79), with cross-validation confirming comparable discriminative performance. Conclusions: Patient-reported measures—particularly quality of life and depressive symptoms—are valuable predictors of hospitalization or death and may enhance traditional frailty and comorbidity assessments in outpatient geriatric care. Future work should focus on developing or integrating screening tools with greater sensitivity to optimize early risk detection and guide preventive interventions. Full article
(This article belongs to the Special Issue Risk Factors for Frailty in Older Adults)
Show Figures

Figure 1

Back to TopTop