{"title":"Kyoto Classification-Based Predictive Factors Associated with the Development of Gastric Cancer After <i>Helicobacter pylori</i> Eradication: A Prospective Multicenter Observational Study.","authors":"Shun Takayama, Osamu Dohi, Ryusuke Horie, Takeshi Yasuda, Tomoko Ochiai, Naoto Iwai, Eiko Imamoto, Tomohisa Takagi, Osamu Handa, Hideyuki Konishi, Takashi Ando, Yuji Naito, Toshiki Takemura, Yoshito Itoh","doi":"10.3390/diagnostics15182376","DOIUrl":"10.3390/diagnostics15182376","url":null,"abstract":"<p><p><b>Background/Objectives</b>: This study aimed to identify specific endoscopic findings associated with the development of GC following successful <i>H. pylori</i> eradication. <b>Methods</b>: This prospective multicenter observational study included patients who underwent annual surveillance endoscopy after successful <i>H. pylori</i> eradication therapy between September 2013 and June 2019. Endoscopic findings were evaluated one year after eradication therapy and analyzed using the Kyoto Classification of Gastritis to identify factors associated with GC development. <b>Results</b>: A total of 465 patients were included, including 49 patients with GC and 416 patients without GC. At the initial endoscopic assessment (median, 0.96 years post-eradication), emergence of map-like redness and invisible regular arrangement of collecting venule (RAC) as independent predictors of GC (map-like redness: hazard ratio [HR], 2.561; 95% confidence interval [CI], 1.362-4.572; <i>p</i> = 0.003; invisible RAC: HR, 3.131; 95% CI, 1.078-9.091; <i>p</i> = 0.036). Patients with map-like redness or invisible RAC showed a significantly higher incidence of GC than those without map-like redness or invisible RAC (<i>p</i> < 0.001 and <i>p</i> < 0.001, respectively). Notably, map-like redness and visible RAC appeared in 13% and 28.4% of cases within the first year after eradication, respectively. <b>Conclusions</b>: Map-like redness and invisible RAC were identified as independent predictors of GC following <i>H. pylori</i> eradication and may serve as early predictive indicators, appearing within one year of successful eradication. This finding underscores the importance of early surveillance endoscopy in identifying patients at elevated risk for GC.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 18","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12468032/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145173883","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-09-18DOI: 10.3390/diagnostics15182381
Siyami Aydın, Mehmet Ağar, Muharrem Çakmak, Mesut Toğaçar
{"title":"Diagnosis of Mesothelioma Using Image Segmentation and Class-Based Deep Feature Transformations.","authors":"Siyami Aydın, Mehmet Ağar, Muharrem Çakmak, Mesut Toğaçar","doi":"10.3390/diagnostics15182381","DOIUrl":"10.3390/diagnostics15182381","url":null,"abstract":"<p><p><b>Background/Objectives</b>: Mesothelioma is a rare and aggressive form of cancer that primarily affects the lining of the lungs, abdomen, or heart. It typically arises from exposure to asbestos and is often diagnosed at advanced stages. Limited datasets and complex tissue structures contribute to delays in diagnosis. This study aims to develop a novel hybrid model to improve the accuracy and timeliness of mesothelioma diagnosis. <b>Methods</b>: The proposed approach integrates automatic image segmentation, transformer-based model training, class-based feature extraction, and image transformation techniques. Initially, CT images were processed using the segment anything model (SAM) for region-focused segmentation. These segmented images were then used to train transformer models (CaiT and PVT) to extract class/type-specific features. Each class-based feature set was transformed into an image using Decoder, GAN, and NeRV techniques. Discriminative score and class centroid analysis were then applied to select the most informative image representation for each input. Finally, classification was performed using a residual-based support vector machine (SVM). <b>Results</b>: The proposed hybrid method achieved a classification accuracy of 99.80% in diagnosing mesothelioma, demonstrating its effectiveness in handling limited data and complex tissue characteristics. <b>Conclusions</b>: The results indicate that the proposed model offers a highly accurate and efficient approach to mesothelioma diagnosis. By leveraging advanced segmentation, feature extraction, and representation techniques, it effectively addresses the major challenges associated with early and precise detection of mesothelioma.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 18","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12468965/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145173938","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-09-18DOI: 10.3390/diagnostics15182379
Ozgen Safak, Mehmet Tolga Hekim, Tolga Cakmak, Fatih Demir, Kursat Demir
{"title":"Improving the Detection Performance of Cardiovascular Diseases from Heart Sound Signals with a New Deep Learning-Based Approach.","authors":"Ozgen Safak, Mehmet Tolga Hekim, Tolga Cakmak, Fatih Demir, Kursat Demir","doi":"10.3390/diagnostics15182379","DOIUrl":"10.3390/diagnostics15182379","url":null,"abstract":"<p><p><b>Background/Objectives</b>: Cardiovascular diseases are among the leading causes of death worldwide. Early diagnosis of these conditions minimizes the risk of future death. Listening to heart sounds with a stethoscope is one of the easiest and fastest methods for diagnosing heart conditions. While heart sounds are a quick and easy diagnostic method, they require significant expert interpretation. Recently, artificial intelligence models trained based on these expert interpretations have become popular in the development of decision support systems. <b>Methods</b>: The proposed approach uses the popular 2016 PhysioNet/CinC Challenge dataset for PCG signals. Spectrogram image transformation was then performed to increase the representativeness of these signals. A deep learning-based model that allows for the simultaneous training of residual and attention blocks and the MLP-mixer model was used for feature extraction. A new algorithm combining the strengths of NCA and ReliefF algorithms was proposed to select the strongest features in the feature set. The SVM algorithm was used for classification. <b>Results</b>: With this proposed approach, over 98% success was achieved in all accuracy, sensitivity, specificity, precision, and F1-score metrics. <b>Conclusions</b>: As a result, an artificial intelligence-based decision support system that detects cardiovascular diseases with high accuracy is presented.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 18","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12468874/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145173660","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-09-18DOI: 10.3390/diagnostics15182382
Fanni Hegedűs, Zsuzsanna Ujfaludi, Orsolya Oláh-Németh, Tamás Lantos, Sándor Turkevi-Nagy, István Balázs Németh, Anita Sejben
{"title":"Nevus with Intralymphatic Nevus Cell Protrusion and Lymphatic Invasion.","authors":"Fanni Hegedűs, Zsuzsanna Ujfaludi, Orsolya Oláh-Németh, Tamás Lantos, Sándor Turkevi-Nagy, István Balázs Németh, Anita Sejben","doi":"10.3390/diagnostics15182382","DOIUrl":"10.3390/diagnostics15182382","url":null,"abstract":"<p><p>We hereby present a case of a 51-year-old woman with a pigmented nodule in the right axillary region. Histopathological examination revealed features consistent with an intradermal nevus. Notably, adjacent to the nevus, intralymphatic protrusion and lymphatic invasion were observed, comprising cells with morphological and immunohistochemical characteristics consistent with nevus cells. Next-generation sequencing revealed the <i>BRAF V600E</i> mutation. To date, 26 similar cases involving intralymphatic nevus cell protrusion and lymphatic invasion have been reported in the literature. Although this finding is rare and may pose a diagnostic challenge for pathologists, it should not be interpreted as indicative of malignancy. Rather, it must be assessed in the context of the lesion's overall histological architecture.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 18","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12468239/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145173914","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-09-18DOI: 10.3390/diagnostics15182368
Fatma Sumer, Merve Yazici
{"title":"Alterations in Corneal Morphology and Thickness Associated with Methylphenidate Treatment in Children with Attention-Deficit/Hyperactivity Disorder.","authors":"Fatma Sumer, Merve Yazici","doi":"10.3390/diagnostics15182368","DOIUrl":"10.3390/diagnostics15182368","url":null,"abstract":"<p><p><b>Background/Objectives</b>: Although methylphenidate is a first-line pharmacological agent in the treatment of Attention-Deficit/Hyperactivity Disorder (ADHD), its long-term effects on ocular tissues, particularly the corneal endothelium, remain poorly understood. Given the cornea's metabolic sensitivity, subclinical changes may occur even in the absence of overt ophthalmologic symptoms. This study aims to evaluate the impact of six-month methylphenidate treatment on corneal endothelial morphology and intraocular pressure (IOP) in pediatric patients with ADHD. <b>Methods</b>: This prospective observational study included 100 treatment-naive children with ADHD and 100 age- and sex-matched healthy controls. All participants underwent comprehensive ophthalmologic assessment at baseline. In the ADHD group, follow-up evaluations were performed after six months of methylphenidate therapy. Endothelial cell density (ECD), average cell area (AVE), standard deviation (SD), coefficient of variation (CV), hexagonality index (6A), central corneal thickness (CCT), and IOP were measured using specular microscopy and corneal topography. ADHD symptom severity was evaluated using the Turgay DSM-IV-Based Rating Scale. <b>Results</b>: Significant reductions in ECD and increases in CCT, CV, AVE, and SD were observed following treatment (<i>p</i> < 0.001). IOP also showed a statistically significant increase while remaining within normal physiological limits. Weak but significant correlations were found between inattention scores and ECD (r = 0.222), and between inattention and corneal volume (r = -0.248). <b>Conclusions</b>: Chronic methylphenidate use may be associated with measurable changes in corneal endothelial microstructure and IOP in children with ADHD. These findings highlight the need for routine ophthalmologic monitoring during stimulant therapy and underscore the importance of further large-scale, long-term studies exploring the neuro-ophthalmologic implications of pediatric psychopharmacological treatment.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 18","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12468723/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145173916","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-09-18DOI: 10.3390/diagnostics15182380
Seda Sağoğlu, Mücahid Yıldırım
{"title":"The Effect of Using a Smartphone App on Oral Hygiene and Brushing Training During Fixed Orthodontic Therapy: A Randomized Clinical Trial.","authors":"Seda Sağoğlu, Mücahid Yıldırım","doi":"10.3390/diagnostics15182380","DOIUrl":"10.3390/diagnostics15182380","url":null,"abstract":"<p><p><b>Objective:</b> This study aimed to study the effectiveness of a smartphone application compared to traditional verbal motivation in improving oral hygiene among fixed orthodontic patients. <b>Methods:</b> Sixty patients were categorized by oral hygiene status using the simplified oral hygiene index (OHI-S) and randomly assigned to either the Dentabuddy group (smartphone application) or the assistant-based training (ABT) group (conventional oral hygiene motivation). Gingival index (GI), plaque index (PI), and gingival bleeding index (GBI) values were recorded at baseline, one month, and three months. Toothbrushing technique was assessed at the three-month follow-up. <b>Results:</b> After three months, the Dentabuddy group exhibited significant GI reductions in participants with fair and poor oral hygiene, whereas the ABT group improved only in those with poor hygiene (<i>p</i> < 0.05). PI values decreased significantly in both groups, except in the ABT group with good and fair hygiene. GBI values improved in both groups, except in the ABT group with fair and poor hygiene (<i>p</i> < 0.05). Toothbrushing demonstrations showed superior technique in the Dentabuddy group (<i>p</i> < 0.05). <b>Conclusions:</b> The Dentabuddy application positively influenced oral hygiene, particularly in individuals with fair and poor hygiene, compared to ABT. This study underscores the potential of smartphone applications in enhancing periodontal health outcomes beyond traditional oral hygiene methods in orthodontic patients with fair or poor hygiene.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 18","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12469057/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145174065","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-09-18DOI: 10.3390/diagnostics15182377
Iuliu-Gabriel Cocuz, Martin Manole, Maria-Cătălina Popelea, Raluca Niculescu, Maria Elena Cocuz, Adrian Horațiu Sabău, Andreea Cătălina Tinca, Andreea Raluca Cozac-Szőke, Diana Maria Chiorean, Alexandru Constantin Ioniță, Eugenia Corina Budin, Georgian-Nicolae Radu, Emoke Andrea Szasz, Ovidiu Simion Cotoi
{"title":"The Epidemiological and Histopathological Profiling of Basal Cell Carcinoma: Insights from a 4-Year Institutional Cohort in a Romanian Clinical County Hospital.","authors":"Iuliu-Gabriel Cocuz, Martin Manole, Maria-Cătălina Popelea, Raluca Niculescu, Maria Elena Cocuz, Adrian Horațiu Sabău, Andreea Cătălina Tinca, Andreea Raluca Cozac-Szőke, Diana Maria Chiorean, Alexandru Constantin Ioniță, Eugenia Corina Budin, Georgian-Nicolae Radu, Emoke Andrea Szasz, Ovidiu Simion Cotoi","doi":"10.3390/diagnostics15182377","DOIUrl":"10.3390/diagnostics15182377","url":null,"abstract":"<p><p><b>Background/Objectives</b>: Basal cell carcinoma (BCC) of the skin is a type of non-melanocytic skin cancer. The European incidence of non-melanocytic skin cancers is 14.2 per 100,000 people, with a mortality rate of 0.5, thus ranking Europe third in the world in terms of incidence and mortality rate, according to the WHO Global Cancer Observatory. The objective of this study was to highlight the histological, epidemiological, and clinicopathological aspects of BCCs diagnosed in the Clinical Pathology Department of the Mures Clinical County Hospital between January 2021 and December 2024. <b>Methods</b>: We performed a retrospective, descriptive, observational study between January 2021 and December 2024 in the Mureș Clinical County Hospital, Targu Mureș, Romania, by analysing data from histopathological reports and histological slides from patients with a positive diagnosis of BCC. The inclusion criteria for this study consisted of patients who presented a histopathological diagnosis of BCCs during the study period. Lesions were divided into two study cohorts-a general cohort and head and neck cohort. The collected data included epidemiological data, macroscopic features, and microscopical characteristics. <b>Results</b>: A total of 540 lesions were included in this study (general cohort), of which 395 were included in the head and neck cohort. This study revealed a higher incidence of BCC in 2024, affecting mostly urban patients (<i>p</i> < 0.001), with more aggressive forms (<i>p</i> < 0.001). The tumours found among males (<i>p</i> = 0.0189) and in rural patients (<i>p</i> = 0.0126) were bigger, but the tumoural volumes decreased over time (<i>p</i> < 0.001). The mixed form of BCC was associated with more aggressive histological subtypes (<i>p</i> < 0.001). <b>Conclusions</b>: BCC presents variability depending on age, gender, environment of origin, and topography, as well as histological subtype and aggressiveness, thus highlighting the need for a personalised approach in terms of diagnostics and treatment.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 18","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12469015/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145174071","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-09-17DOI: 10.3390/diagnostics15182366
Emiliano Fiori, Sara Corradetti, Giovanna Gallo, Alberto Palazzuoli, Antonio Pagliaro, Roberta Molle, Pier Giorgio Tiberi, Elisabetta Salvioni, Arianna Piotti, Paola Gugliandolo, Piergiuseppe Agostoni, Damiano Magrì, Emanuele Barbato
{"title":"Clinical and Prognostic Impact of Hemodynamic Gain Index and Heart Hemodynamic Reserve in Heart Failure with Reduced and Mildly Reduced Ejection Fraction: A Multicenter Study.","authors":"Emiliano Fiori, Sara Corradetti, Giovanna Gallo, Alberto Palazzuoli, Antonio Pagliaro, Roberta Molle, Pier Giorgio Tiberi, Elisabetta Salvioni, Arianna Piotti, Paola Gugliandolo, Piergiuseppe Agostoni, Damiano Magrì, Emanuele Barbato","doi":"10.3390/diagnostics15182366","DOIUrl":"10.3390/diagnostics15182366","url":null,"abstract":"<p><p><b>Background/Objectives:</b> Cardiopulmonary exercise testing (CPET) is a well-established tool for risk stratification in patients with heart failure (HF); however, its utility is limited in routine clinical practice due to the associated cost and technical demands. The hemodynamic gain index (HGI), a non-metabolic parameter derived from systolic blood pressure and heart rate changes during exercise, has been demonstrated to play a promising role in HF populations. In this study, we aimed both to validate the prognostic value of the HGI and to evaluate a novel metric, heart hemodynamic reserve (HHR), in patients with HF and left ventricular ejection fraction (LVEF) below 50%. <b>Methods:</b> We retrospectively enrolled 479 consecutive patients with HF and reduced or mildly reduced LVEF who underwent maximal, symptom-limited CPET at three Italian university hospitals between 2012 and 2024. The HGI and HHR were computed using resting and peak exercise hemodynamic data. HHR is defined as the product of systolic blood pressure and heart rate reserve with exercise, normalized for the age-predicted maximum heart rate. The primary endpoint was a composite of cardiovascular death, urgent heart transplantation (HTx), or left ventricular assist device (LVAD) implantation. Prognostic associations were assessed using multivariable Cox regression and area under the receiver operating characteristic curves (AUCs). <b>Results:</b> During a median follow-up of 3.25 years, the composite outcome occurred in 56 patients (11.5%). Both the HGI and HHR were independently associated with the prespecified endpoint (HGI HR: 0.41, 95% CI: 0.20-0.83, <i>p</i> = 0.013; HHR HR: 0.89, 95% CI: 0.83-0.96, <i>p</i> = 0.004), with HHR showing a slightly higher prognostic accuracy than the HGI (AUC 0.78 vs. 0.74; <i>p</i> = 0.033). <b>Conclusions:</b> Both the HGI and HHR are independent prognostic markers in HF patients with LVEF < 50%. Their non-metabolic derivation makes them valuable tools for risk stratification in settings where CPET is unavailable.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 18","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12468042/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145174073","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-09-17DOI: 10.3390/diagnostics15182362
Adi Alhudhaif
{"title":"Pain Level Classification from Speech Using GRU-Mixer Architecture with Log-Mel Spectrogram Features.","authors":"Adi Alhudhaif","doi":"10.3390/diagnostics15182362","DOIUrl":"10.3390/diagnostics15182362","url":null,"abstract":"<p><p><b>Background/Objectives</b>: Automatic pain detection from speech signals holds strong promise for non-invasive and real-time assessment in clinical and caregiving settings, particularly for populations with limited capacity for self-report. <b>Methods:</b> In this study, we introduce a lightweight recurrent deep learning approach, namely the Gated Recurrent Unit (GRU)-Mixer model for pain level classification based on speech signals. The proposed model maps raw audio inputs into Log-Mel spectrogram features, which are passed through a stacked bidirectional GRU for modeling the spectral and temporal dynamics of vocal expressions. To extract compact utterance-level embeddings, an adaptive average pooling-based temporal mixing mechanism is applied over the GRU outputs, followed by a fully connected classification head alongside dropout regularization. This architecture is used for several supervised classification tasks, including binary (pain/non-pain), graded intensity (mild, moderate, severe), and thermal-state (cold/warm) classification. End-to-end training is done using speaker-independent splits and class-balanced loss to promote generalization and discourage bias. The provided audio inputs are normalized to a consistent 3-s window and resampled at 8 kHz for consistency and computational efficiency. <b>Results:</b> Experiments on the TAME Pain dataset showcase strong classification performance, achieving 83.86% accuracy for binary pain detection and as high as 75.36% for multiclass pain intensity classification. <b>Conclusions:</b> As the first deep learning based classification work on the TAME Pain dataset, this work introduces the GRU-Mixer as an effective benchmark architecture for future studies on speech-based pain recognition and affective computing.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 18","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12468818/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145173895","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-09-17DOI: 10.3390/diagnostics15182360
Ivana Lapić, Dragana Šegulja, Željkica Jakoplić, Iva Lukić, Dunja Rogić
{"title":"Reference Intervals and Cut-Off Values for Thyroid Tests in the Croatian Adult Population on the Snibe MAGLUMI X6 Immunoassay Analyzer.","authors":"Ivana Lapić, Dragana Šegulja, Željkica Jakoplić, Iva Lukić, Dunja Rogić","doi":"10.3390/diagnostics15182360","DOIUrl":"10.3390/diagnostics15182360","url":null,"abstract":"<p><p><b>Background/Objectives:</b> To establish reference intervals (RIs) and cut-off values for thyroid-related tests on the MAGLUMI X6 immunoassay analyzer (Snibe Diagnostic, Shenzhen, China) in an adult Croatian population. <b>Methods</b>: This study included 305 healthy individuals who underwent regular preventive medical checkup. The following tests were determined in serum: thyroid-stimulating hormone (TSH), total triiodothyronine (TT3), total thyroxine (TT4), free triiodothyronine (FT3), free thyroxine (FT4), thyroglobulin (Tg), reverse triiodothyronine (revT3), total binding capacity of thyroglobulin (T-uptake), thyroglobulin antibodies (anti-Tg), anti-thyroid peroxidase antibodies (anti-TPO) and thyroid receptor antibodies (TRAb). TSH, TT3, TT4, FT3, FT4, Tg, revT3 and T-uptake results were used for calculating double-sided 95% RIs between the 2.5th and 97.5th percentiles. For anti-Tg, anti-TPO and TRAb, right-sided cut-offs that correspond to the 95th percentile were determined. <b>Results</b>: Reference intervals for TSH, TT4, FT3, FT4, Tg, T-uptake and revT3 did not differ by gender (<i>p</i> > 0.05) and were 0.77-5.04 mIU/L, 69.9-127.7 nmol/L, 3.84-6.20 pmol/L, 13.8-19.7 pmol/L, 1.8-51.2 µg/L, 0.9-1.2 TBI and 0.44-0.73 ng/mL, respectively. The RI for TT3 was different for males (1.49-2.53 nmol/L) and females (1.43-2.81 nmol/L), <i>p</i> = 0.021. A single cut-off for anti-TPO was established (<18 kIU/L). Differences in cut-offs for males and females were obtained for anti-Tg (<72 and <104 kIU/L, respectively) and TRAb (0.6 and 0.9 IU/L, respectively). <b>Conclusions</b>: This is the first study to determine RIs for thyroid function tests in Croatian adults on the Snibe analytical platform. The obtained results point out to the use of population- and immunoassay-specific RIs. For TT3, anti-Tg and TRAb gender-specific RIs should be considered.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 18","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12468744/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145173930","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}