Diagnostics

18 pages, 980 KB

Open AccessArticle

IDF-Net: Interpretable Dynamic Fusion Network for Colorectal Cancer Diagnosis Using Cross-Modal Imaging

by Helen Haile Hayeso, Peifeng Shi, Jingwen Lian, Zenebe Markos Lonseko and Nini Rao

Diagnostics 2026, 16(1), 99; https://doi.org/10.3390/diagnostics16010099 (registering DOI) - 27 Dec 2025

Background/Objectives: Colorectal cancer (CRC) is a leading cause of cancer deaths worldwide, underscoring the need for diagnostic tools that early, accurate, and clinically interpretable. Current artificial intelligence (AI) models are predominantly unimodal and lack sufficient interpretability, which restricts their clinical adoption. Methods [...] Read more.

Background/Objectives: Colorectal cancer (CRC) is a leading cause of cancer deaths worldwide, underscoring the need for diagnostic tools that early, accurate, and clinically interpretable. Current artificial intelligence (AI) models are predominantly unimodal and lack sufficient interpretability, which restricts their clinical adoption. Methods: We propose IDF-Net, an interpretable dynamic fusion framework that integrates endoscopy, computed tomography (CT), and histopathology using modality-specific encoders, a dual-stage adaptive gating mechanism, and cross-modal attention. We conducted stratified 5-fold cross-validation and assessed interpretability using spatial heatmaps and modality attribution. We also quantified the results using the intersection-over-union metric for saliency alignment. Results: IDF-Net achieved a state-of-the-art accuracy of 0.920 (0.907–0.936) and area under the curve (AUC) of 0.991 (95% CI: 0.965–0.997), significantly outperforming unimodal and static-fusion baselines (p < 0.05). Interpretability analysis of IDF-Net demonstrated a strong alignment between Gradient-weighted Class Activation Mapping++ heatmaps and expert-annotated lesions, as well as case-specific modality contributions via SHapley Additive exPlanations values. Ablation studies confirmed the contribution of each component, with dynamic routing and cross-attention fusion improving AUC by 0.038 and 0.046, respectively. Conclusions: IDF-Net introduces a dynamically fused, multimodal diagnostic framework with integrated quantitative interpretability, demonstrating superior accuracy and strong potential for clinical translation in CRC diagnosis. The model’s adaptive design allows it to function robustly even when CT data is unavailable, aligning with common clinical pathways while leveraging additional imaging when present for comprehensive staging. Full article

(This article belongs to the Special Issue Artificial Intelligence in Gastrointestinal Disease: Diagnosis and Management in 2025)

22 pages, 690 KB

Open AccessReview

Innovations in Diagnosis and Treatment of Coronary Artery Disease

by Salaheldin Agamy, Sheref Zaghloul, Zahid Khan, Ahmed Shahin, Ramy Kishk, Ahmed Smman and Luciano Candilio

Diagnostics 2026, 16(1), 98; https://doi.org/10.3390/diagnostics16010098 (registering DOI) - 27 Dec 2025

Abstract

Background: Coronary artery disease (CAD) remains a significant health challenge, placing a heavy burden on people and healthcare systems worldwide. Objectives: This narrative review aims to provide a comprehensive overview of recent advancements in the diagnosis, intervention, and pharmacological management of [...] Read more.

Background: Coronary artery disease (CAD) remains a significant health challenge, placing a heavy burden on people and healthcare systems worldwide. Objectives: This narrative review aims to provide a comprehensive overview of recent advancements in the diagnosis, intervention, and pharmacological management of CAD, with a focus on emerging technologies shaping its future. Methods: This is a narrative review that synthesises information from diverse sources, including clinical trials, systematic reviews, meta-analyses, and preclinical studies, to provide a comprehensive overview of the current landscape and emerging trends in CAD management. The literature included in this review was sourced from original research articles and review papers published between January 2010 and December 2025. Results: Early detection has been transformed by non-invasive imaging, such as PCAT, and the addition of invasive and non-invasive FFR technology enables quicker and more accurate diagnoses. Biomarkers, such as high-sensitivity troponin, have further improved the precision of acute coronary syndrome detection, enhancing early intervention. In interventional cardiology, new-generation drug-eluting stents (DESs) have lowered restenosis rates, whereas robotic-assisted percutaneous coronary intervention (PCI) offers precision and reduced operator radiation exposure. Furthermore, the efficacy of drug-coated balloons (DCBs) has been established in the management of in-stent restenosis, and their application in de novo coronary lesions and bifurcation anatomy remains promising. Looking ahead, nanomedicine promises targeted plaque reduction and vascular repair, while 3D-bioprinted blood vessels offer durable, biocompatible grafts for surgical applications. Pharmacological developments, including modern cholesterol-lowering drugs, have also been crucial in achieving cholesterol targets. Conclusions: Despite significant advancements in diagnosis, intervention, and pharmacotherapy, several critical challenges remain, including the need for validated biomarkers and imaging modalities to identify vulnerable atheroma before symptoms arise. Continued research is essential to improve patient outcomes and address the global burden of CAD. Full article

(This article belongs to the Special Issue Diagnosis and Management of Coronary Heart Disease)

► Show Figures

Figure 1

24 pages, 1571 KB

Open AccessArticle

Quantifying Explainability in OCT Segmentation of Macular Holes and Cysts: A SHAP-Based Coverage and Factor Contribution Analysis

by İlknur Tuncer Fırat, Murat Fırat and Taner Tuncer

Diagnostics 2026, 16(1), 97; https://doi.org/10.3390/diagnostics16010097 (registering DOI) - 27 Dec 2025

Abstract

Background: Optical coherence tomography (OCT) can quantify the morphology and dimensions of a macular hole for diagnosis and treatment planning. Objective: The aim of this study was to perform automatic segmentation of macular holes (MHs) and cysts from OCT macular volumes using [...] Read more.

Background: Optical coherence tomography (OCT) can quantify the morphology and dimensions of a macular hole for diagnosis and treatment planning. Objective: The aim of this study was to perform automatic segmentation of macular holes (MHs) and cysts from OCT macular volumes using a deep learning-based model and to quantitatively evaluate decision reliability using the model’s focus regions and GradientSHAP-based explainability. Methods: In this study, we automatically segmented MHs and cysts in OCT images from the open-access OIMHS dataset. The dataset comprises 125 eyes from 119 patients and 3859 OCT B-scans. OCT B-scan slices were input to a UNet-48-based model with a 2.5D stacking strategy. Performance was evaluated using Dice and intersection-over-union (IoU), boundary accuracy was evaluated using the 95th-percentile Hausdorff distance (HD95), and calibration was evaluated using the expected calibration error (ECE). Explainability was quantified from GradientSHAP maps using lesion coverage and spatial focus metrics: Attribution Precision in Lesion (APILτ), which is the proportion of attributions (SHAP contributions) falling inside the lesion; Attribution Recall in Lesion (ARILτ), which is the proportion of the true lesion covered by the attributions; and leakage (Leakτ = 1 − APILτ), which is the proportion of attributions falling outside the lesion. Spatial focus was monitored using the center-of-mass distance (COM-dist), which is the Euclidean distance between the attribution center and the segmentation center. All metrics were calculated using the top τ% of the pixels with the highest SHAP values. SHAP features were clustered using PCA and k-means. Explanations were calculated using the clinical mask in ground truth (GT) mode and the model segmentation in prediction (Pred) mode. Results: The Dice/IoU values for holes and cysts were 0.94/0.91 and 0.87/0.81, respectively. Across lesion classes, HD95 = 6 px and ECE = 0.008, indicating good boundary accuracy and calibration. In GT mode (τ = 20), three regimes were observed: (i) retina-dominant: high ARIL (hole: 0.659; cyst: 0.654), high Leak (hole: 0.983; cyst: 0.988), and low COM-dist (hole: 7.84 px; cyst: 6.91 px), with the focus lying within the retina and largely confined to the retinal tissue; (ii) peri-lesional: highest ARIL (hole: 0.684; cyst: 0.719), relatively lower Leak (hole: 0.917; cyst: 0.940), and medium/high COM-dist (hole: 16.22 px; cyst: 10.17 px), with the focus located around the lesion; (iii) narrow-coverage: primarily seen for cysts in GT mode (ARIL: 0.494; Leak: 1.000; COM-dist: 52.02 px), with markedly reduced coverage. In Pred mode, the ARIL20 for holes increased in the retina-dominant cluster (0.758) and COM-dist decreased (6.24 px), indicating better agreement with the model segmentation. Conclusions: The model exhibited high accuracy and good calibration for MH and cyst segmentation in OCT images. Quantitative characterization of SHAP validated the model results. In the clinic, peri-lesion and narrow-coverage conditions are the key situations that require careful interpretation. Full article

(This article belongs to the Topic Machine Learning and Deep Learning in Medical Imaging)

20 pages, 505 KB

Open AccessArticle

Comparative Fractal Analysis on Orthopantomography Images of Bone Remodeling Changes at 3 Months Between Natural Socket Healing and Immediate Implant Placement

by Andrei Radu, Mihaela Ionescu, Antonia Samia Khaddour, Cristina Maria Munteanu, Monica Mihaela Iacov Crăițoiu, Melania Olimpia Cojocaru, Alex Ioan Sălan, Iulia Roxana Marinescu, Răzvan Mercuț and Sanda Mihaela Popescu

Diagnostics 2026, 16(1), 96; https://doi.org/10.3390/diagnostics16010096 (registering DOI) - 27 Dec 2025

Abstract

Background: Osseointegration is the main factor that ensures the long-term success of implant-prosthetic therapy, but besides this, there are other important factors, such as the quality of the alveolar bone and the time of placement of dental implants. The study aimed to analyze [...] Read more.

Background: Osseointegration is the main factor that ensures the long-term success of implant-prosthetic therapy, but besides this, there are other important factors, such as the quality of the alveolar bone and the time of placement of dental implants. The study aimed to analyze changes in the alveolar bone following tooth extraction, comparing natural healing with immediate implant placement, using fractal analysis on OPG images. Methods: This retrospective study included OPG images obtained before tooth extraction and 3 months after surgery in 91 patients who underwent maxillary and mandibular molar extractions and opted for either natural healing or immediate dental implant placement. Fractal analysis of OPG images was performed using Image J software, and the resulting measurements were subsequently statistically analyzed. Results: Most extractions were performed in the maxilla, and most were at the level of the first molar. The study group showed a faster healing process following immediate placement of dental implants, regardless of location, and a similar distribution of bone resorption and healing, with clear differences in location: the mandible had a faster healing process than the maxilla. Conclusions: Fractal analysis showed a better and quicker bone healing of the alveolar bone in immediate implant placement in molar areas compared with post-extraction natural healing, especially in the lower jaw. Full article

(This article belongs to the Special Issue Application of Diagnostic Tools for Dentistry and Maxillofacial Surgery)

12 pages, 375 KB

Open AccessArticle

Exploring the Role of Cerebrospinal Fluid and Serum Mid-Regional Pro-Adrenomedullin in Tick-Borne Encephalitis: A Pilot Study

by Gabriela Trojan, Anna Moniuszko-Malinowska, Joanna Oklińska, Wioletta Pawlak-Zalewska, Ewelina Kruszewska, Agnieszka Kulczyńska-Przybik, Barbara Mroczko and Piotr Czupryna

Diagnostics 2026, 16(1), 95; https://doi.org/10.3390/diagnostics16010095 (registering DOI) - 27 Dec 2025

Abstract

Background: Adrenomedullin (ADM) is a multifunctional peptide with vasoregulatory, antimicrobial, and anti-inflammatory properties. Its stable fragment, mid-regional pro-adrenomedullin (MR-proADM), is a validated biomarker in sepsis and systemic infections, but its role in viral neuroinfections remains unexplored. Tick-borne encephalitis (TBE), caused by the [...] Read more.

Background: Adrenomedullin (ADM) is a multifunctional peptide with vasoregulatory, antimicrobial, and anti-inflammatory properties. Its stable fragment, mid-regional pro-adrenomedullin (MR-proADM), is a validated biomarker in sepsis and systemic infections, but its role in viral neuroinfections remains unexplored. Tick-borne encephalitis (TBE), caused by the tick-borne encephalitis virus (TBEV), is a major viral infection of the central nervous system (CNS) associated with long-term neurological sequelae. This study aimed to assess MR-proADM levels in cerebrospinal fluid (CSF) and serum of patients with TBE and to evaluate their diagnostic utility and pathophysiological significance. Methods: This retrospective observational study included 20 patients with confirmed TBE and 14 non-infectious neurological controls. MR-proADM concentrations were measured in paired CSF and serum samples using an ELISA assay. Statistical analyses included group comparisons (Mann–Whitney U test), correlation analyses (Spearman’s r), and receiver operating characteristic (ROC) curve evaluation. Results: Serum MR-proADM levels at baseline (SER1) were significantly lower in TBE patients compared with controls (p = 0.0197). The CSF/serum MR-proADM ratio differed significantly between groups (p = 0.0063) and showed the best diagnostic performance (AUC = 0.816, 95% CI 0.63–0.93; sensitivity 79%, specificity 80%). MR-proADM concentrations in CSF correlated with total CSF protein (r = 0.53), suggesting an association with blood–CSF barrier dysfunction. Strong reproducibility was observed for serum MR-proADM between sampling points (r = 0.83). Conclusions: MR-proADM levels in CSF and serum are altered in patients with TBE, indicating its potential as a biomarker of CNS infection and inflammation. The CSF/serum MR-proADM ratio may serve as a sensitive indicator of blood–CSF barrier involvement, while decreased serum levels may reflect impaired systemic neuroprotective response. These findings highlight a possible role of ADM in neuroimmune regulation during viral encephalitis and warrant validation in larger prospective studies. Full article

(This article belongs to the Special Issue Advances in the Diagnosis and Treatment of Infections of the Central Nervous System)

► Show Figures

Figure 1

23 pages, 638 KB

Open AccessReview

Acute Kidney Injury Biomarkers in Perioperative Care: A Scoping Review of Clinical Implementation

by Konrad Zuzda, Paulina Walczak-Wieteska, Paweł Andruszkiewicz and Jolanta Małyszko

Diagnostics 2026, 16(1), 94; https://doi.org/10.3390/diagnostics16010094 (registering DOI) - 27 Dec 2025

Abstract

Background: Acute kidney injury (AKI) remains one of the most common perioperative complications, carrying substantial mortality and healthcare burden. Traditional diagnostic criteria relying on serum creatinine and urine output are limited by delayed detection and inability to characterize the underlying injury phenotype. [...] Read more.

Background: Acute kidney injury (AKI) remains one of the most common perioperative complications, carrying substantial mortality and healthcare burden. Traditional diagnostic criteria relying on serum creatinine and urine output are limited by delayed detection and inability to characterize the underlying injury phenotype. This scoping review examined the current state of novel AKI biomarker research in perioperative care, evaluated their clinical implementation, and identified knowledge gaps. Methods: A systematical search was performed for studies investigating novel AKI biomarkers in surgical settings. Biomarkers were categorized as functional, stress, or damage markers. Data extraction focused on diagnostic performance, clinical outcomes, regulatory approval status, and implementation barriers. A narrative synthesis was organized by biomarker category and thematic areas. Results: Several biomarkers demonstrated superior early diagnostic performance compared to traditional ones, including PENK or CCL-14, showing promising accuracy for AKI detection and outcome prediction. TIMP-2*IGFBP-7 and NGAL achieved regulatory approval, and biomarker-guided KDIGO care bundles significantly reduced AKI incidence in surgical populations. However, substantial heterogeneity exists in assays, cutoff values, and clinical validation across different clinical settings. Conclusions: Novel AKI biomarkers offer a promise for early detection and risk stratification in perioperative care, yet widespread clinical adoption requires addressing standardization challenges, establishing cost-effectiveness, and validating implementation strategies. Full article

(This article belongs to the Special Issue Predictive and Prognostic Markers in Critically Ill Patients, 2nd Edition)

► Show Figures

Figure 1

18 pages, 3425 KB

Open AccessArticle

All About Multiparametric MRI Evaluation in Biliary Tree Complications After Liver Transplant

by Adrian Dumitru Dijmărescu, Cristina Dumitrescu, Cristina Alexandra Nicolae, Robert Mihai Enache and Ioana Gabriela Lupescu

Diagnostics 2026, 16(1), 93; https://doi.org/10.3390/diagnostics16010093 (registering DOI) - 27 Dec 2025

Abstract

Background/Objectives: To present, discuss, and illustrate the role of multiparametric magnetic resonance imaging (MPMRI) in the evaluation of biliary tree (BT) complications after liver transplantation (LT) as an integrated part into the multidisciplinary team approach for personalized patients’ treatment. Methods: We [...] Read more.

Background/Objectives: To present, discuss, and illustrate the role of multiparametric magnetic resonance imaging (MPMRI) in the evaluation of biliary tree (BT) complications after liver transplantation (LT) as an integrated part into the multidisciplinary team approach for personalized patients’ treatment. Methods: We retrospectively analyzed the MPMRI findings of 317 patients out of 1080 cases with LT, admitted to the Fundeni Clinical Institute from January 2005 to June 2025, who developed biliary complications. Results: Biliary complications after LT evaluated by MPMRI included anastomotic strictures in 235 cases (74%), intra- or extrahepatic bile leaks/biloma in 56 patients (18%), secondary cholangitis due to pyogenic cholangitis in 91 cases (29%), liver abscesses in 23 patients (7%), BT lithiasis in 27 patients (8.5%), disease recurrence in 26 cases (8%), and extrinsic BT compression in 1 case (0.3%). Conclusions: MPMRI plays a crucial role for the evaluation of BT complications, with the protocol being optimized in correlation with the clinical question or suspicion and with the clinical status of the patient. Full article

(This article belongs to the Special Issue Advancements in the Diagnosis and Management of Liver Disease and Liver Transplantation)

► Show Figures

Figure 1

0 pages, 1029 KB

Open AccessArticle

Resolution Comparison of a Standoff Gel Pad Versus a Liquid Gel Barrier for Nasal Bone Fracture Sonography: A Standardized Crossover Study

by Dong Gyu Kim and Kyung Ah Lee

Diagnostics 2026, 16(1), 92; https://doi.org/10.3390/diagnostics16010092 (registering DOI) - 26 Dec 2025

Abstract

Background: High-frequency ultrasonography (US) is increasingly used to guide closed reduction in nasal bone fractures, but near-field resolution over the curved nasal dorsum depends critically on the acoustic coupling medium. We aimed to determine whether a semi-solid standoff gel pad (PAD) provides [...] Read more.

Background: High-frequency ultrasonography (US) is increasingly used to guide closed reduction in nasal bone fractures, but near-field resolution over the curved nasal dorsum depends critically on the acoustic coupling medium. We aimed to determine whether a semi-solid standoff gel pad (PAD) provides superior image contrast and signal stability compared with a liquid gel barrier (LGB) during intraoperative nasal bone fracture sonography. Methods: In this prospective, single-center, within-subject crossover study, 30 adults with isolated nasal bone fractures underwent intraoperative high-frequency US of the nasal dorsum under two coupling conditions differing only by the medium used: a 7 mm hydrogel standoff pad (PAD) and a custom-made 7 mm liquid gel barrier (LGB). All scans were acquired on the same platform using fixed B-mode presets (10 MHz, 4.0 cm depth, single focal zone at the cortex). Rectangular regions of interest (ROIs) were placed on the cortical interface (bone ROI) and adjacent soft tissue (soft-tissue ROI) at matched depth. For each subject and condition, contrast-to-noise ratio (CNR) and two signal-to-noise ratios (SNR_bone, SNR_soft) were derived from ROI gray-level statistics and compared using paired t-tests. Results: The PAD yielded a significantly higher CNR at the cortical interface compared to the LGB (3.46 ± 0.17 vs. 2.50 ± 0.19; mean paired difference 0.96, 95% CI 0.88–1.04; p < 0.0001). SNR_bone was also higher with PAD (4.31 ± 0.35 vs. 3.63 ± 0.34; difference 0.68, 95% CI 0.52–0.83; p < 0.0001). Using the soft-tissue ROI as the noise reference (SNR_soft), PAD again outperformed LGB (7.64 ± 0.73 vs. 6.68 ± 0.78; difference 0.96, 95% CI 0.59–1.33; p = 0.000012). Conclusions: Compared with a liquid gel barrier of similar thickness, a semi-solid standoff gel pad provides higher near-field CNR and SNR at the nasal cortical interface under standardized intraoperative conditions. These quantitative differences support the use of a gel pad as a practical coupling medium for real-time ultrasound guidance during closed reduction in nasal bone fractures, although the impact on clinical outcomes remains to be determined. Full article

(This article belongs to the Special Issue Advances in Plastic Surgery: Diagnosis, Management and Prognosis)

► Show Figures

Figure 1

11 pages, 242 KB

Open AccessArticle

Utility of the BioFire^® FilmArray^® Pneumonia Panel Using Bronchial Washing Specimens: A Comparative Study with Conventional Culture

by Sungjin Jo, Sei Won Kim, Jung Ok Kim, Sang-hyun Shin, Sehee Kim, Heayon Lee, Chang Dong Yeo, Sang Haak Lee, In Young Yoo and Yeon-Joon Park

Diagnostics 2026, 16(1), 91; https://doi.org/10.3390/diagnostics16010091 (registering DOI) - 26 Dec 2025

Abstract

Background: Accurate identification of pneumonia pathogens is critical for guiding appropriate antibiotic therapy and minimizing unnecessary antimicrobial exposure. Bronchoalveolar lavage (BAL) is widely used for pathogen detection but introduces procedural risks. Bronchial washing (BW) is simpler and less invasive, yet evidence for its [...] Read more.

Background: Accurate identification of pneumonia pathogens is critical for guiding appropriate antibiotic therapy and minimizing unnecessary antimicrobial exposure. Bronchoalveolar lavage (BAL) is widely used for pathogen detection but introduces procedural risks. Bronchial washing (BW) is simpler and less invasive, yet evidence for its utility in multiplex PCR diagnostics is limited. Methods: This study includes an evaluation of the clinical utility of the BioFire^® FilmArray^® Pneumonia Panel (FA) using BW specimens via comparison with conventional culture. Between 2022 and 2024, 190 BW specimens were collected from 182 adult patients with suspected pneumonia at Eunpyeong St. Mary’s Hospital, Seoul, Korea. Each specimen was tested simultaneously using FA and conventional culture. Results: The culture positivity rate was 41.6%, whereas FA showed a higher positivity rate of 51.1%. Of all specimens, 52.6% (100/190) were positive in at least one of two methods, and 11.0% (21/190) were FA-positive only. FA detected 72 additional bacterial targets, most commonly H. influenzae, K. pneumoniae, S. aureus, S. agalactiae, and S. pneumoniae. Semi-quantitative results demonstrated a statistically significant moderate correlation with culture (ρ = 0.48, p < 0.001). Eight bacterial targets achieved 100% PPA, and resistance genes were rapidly detected, although some discrepancies with phenotypic antimicrobial susceptibility testing were observed. Several FA-only detections may reflect oropharyngeal colonization rather than true infection. Conclusions: FA testing of BW specimens demonstrated high concordance with culture and provided rapid pathogen and resistance gene detection. BW-based FA testing may serve as a useful diagnostic alternative when BAL is not feasible, although careful interpretation is required to account for potential contamination. Full article

(This article belongs to the Section Diagnostic Microbiology and Infectious Disease)

14 pages, 2403 KB

Open AccessArticle

Evaluation of CD3 and CD20 Lymphocytes and Mast Cells in the Microenvironment of Central Giant Cell Granuloma, Peripheral Giant Cell Granuloma, and Giant Cell Tumor of Bone

by Khelan A. Fayaq and Balkees T. Gharib

Diagnostics 2026, 16(1), 90; https://doi.org/10.3390/diagnostics16010090 (registering DOI) - 26 Dec 2025

Abstract

Objective: Giant cell lesions (GCLs) share similar histopathologic features. The influence of immune involvement on the biology of giant cell lesions remains largely elusive. This study aimed to evaluate and compare lymphocyte and mast cell infiltration and distribution among three giant cell [...] Read more.

Objective: Giant cell lesions (GCLs) share similar histopathologic features. The influence of immune involvement on the biology of giant cell lesions remains largely elusive. This study aimed to evaluate and compare lymphocyte and mast cell infiltration and distribution among three giant cell lesions. Study design: A total of 30 FFPE tissue blocks, comprising 10 PGCGs, 10 CGCGs (aggressive and nonaggressive), and 10 GCTs (aggressive and nonaggressive) of bone, were subjected to IHC staining for CD3 and CD20 lymphocyte markers and toluidine blue staining for mast cells. The mean count of positively stained cells was calculated and categorized into three scores, along with a group for negative cases. Statistical analysis was conducted to assess significance at p < 0.05. Result: Lymphocyte infiltration was observed across all lesions. CD3⁺ and CD20⁺ cell counts were significantly elevated in PGCGs, followed by CGCGs, and were lowest in GCTs of bone. In contrast, mast cell counts were high in GCTs of bone and CGCGs and low in PGCGs. Aggressive giant cell lesions of bone showed a significantly low number of CD3⁺ and CD20⁺ cells (Mann–Whitney U test; p = 0.05, 0.004) and a high number of mast cells (Mann–Whitney U test; p < 0.001) compared with nonaggressive lesions of bone. PGCGs and nonaggressive CGCGs showed comparable CD3 expression, with no significant difference between them (p = 0.59). CD20 levels were higher in nonaggressive CGCGs but did not reach statistical significance (Mann–Whitney U test; p = 0.07). Mast cell density was significantly lower in PGCGs compared with intraosseous nonaggressive CGCGs. Conclusions: The present study shows that GCTs of bone, CGCGs, and PGCGs possess distinct immune microenvironmental profiles. Aggressive lesions demonstrate reduced lymphocyte infiltration and increased mast cell density, a pattern particularly evident in GCTs of bone. This imbalance may contribute to their aggressive behavior by enabling them to escape host immune regulation. Full article

(This article belongs to the Section Pathology and Molecular Diagnostics)

► Show Figures

Figure 1

17 pages, 1393 KB

Open AccessArticle

Cellular Immunological Memory T Cells and IL15RA Gene Polymorphism in COVID-19 Vaccinated Individuals from Southern Brazil

by Grazielle Motta Rodrigues, Pâmela Portela da Silva, Maria Clara de Freitas Pinho, Taís da Silveira Fischer, Fernanda de Paris, Fabrício Souza Campos, Arthur Bandeira de Mello Garcia, Lucas Fernandes Jataí, Patricia Ashton-Prolla, Fernanda Sales Luiz Vianna and Clévia Rosset

Diagnostics 2026, 16(1), 89; https://doi.org/10.3390/diagnostics16010089 (registering DOI) - 26 Dec 2025

Abstract

Background: The development of safe and effective vaccines against SARS-CoV-2 was crucial for controlling COVID-19 and establishing long-lasting immune memory in the population. Methods: This study evaluated cellular immune memory in individuals vaccinated with different regimens in Rio Grande do Sul using flow [...] Read more.

Background: The development of safe and effective vaccines against SARS-CoV-2 was crucial for controlling COVID-19 and establishing long-lasting immune memory in the population. Methods: This study evaluated cellular immune memory in individuals vaccinated with different regimens in Rio Grande do Sul using flow cytometry. Additionally, the rs2228059 polymorphism in the IL15RA gene was genotyped. A total of 62 participants were randomly recruited. Results: A decrease in memory T cell subsets in response to SARS-CoV-2 stimuli was observed in total CD3⁺, CD4⁺, and CD8⁺ T cells. Regarding the timing of the last vaccine dose, 94.4% of participants had received their final COVID-19 vaccination at least two years prior to recruitment. The rs2228059 polymorphism was genotyped in 443 individuals from the Rio Grande do Sul population. Among participants who received the ChAdOx1/ChAdOx1/BNT162b2 vaccination regimen and carried the TT genotype, an increase in CD8⁺ naive, CD8⁺ effector and CD4⁺ naive subsets was observed in stimulated cells. Although preliminary, the results suggest no major differences between vaccination regimens. Conclusions: The progressive reduction in memory T cell counts supports the need for booster doses, which is essential not only in the context of new emerging variants but also especially to maintain adequate cellular immune protection. Full article

(This article belongs to the Special Issue Opportunities in Laboratory Medicine in the Era of Genetic Testing)

► Show Figures

Figure 1

16 pages, 5774 KB

Open AccessArticle

Hyperuricemia-Informed Survival Machine-Learning Prediction of Post-Thrombotic Syndrome After Unprovoked DVT: A Dual-Center Prospective Study

by Yajing Li, Hongru Deng and Yongquan Gu

Diagnostics 2026, 16(1), 88; https://doi.org/10.3390/diagnostics16010088 (registering DOI) - 26 Dec 2025

Abstract

Background/Objectives: Post-thrombotic syndrome (PTS) following unprovoked deep vein thrombosis (DVT) lacks readily available, calibrated risk estimates at defined follow-up horizons. Building on signals that thrombus burden, care processes, and a form of metabolic–inflammatory tone influence outcomes, we prospectively evaluated survival machine-learning models, [...] Read more.

Background/Objectives: Post-thrombotic syndrome (PTS) following unprovoked deep vein thrombosis (DVT) lacks readily available, calibrated risk estimates at defined follow-up horizons. Building on signals that thrombus burden, care processes, and a form of metabolic–inflammatory tone influence outcomes, we prospectively evaluated survival machine-learning models, explicitly including hyperuricemia while excluding what we consider major inflammatory confounders. Methods: Adults with first-episode unprovoked lower-extremity DVT were enrolled at two centers (July 2024–September 2025). PTS (Villalta) was assessed at 3, 6, 9, and 12 months. The cohort was split 70/30 into training and test sets. Eight learners (RSF, GBM, LASSO + Cox, CoxBoost, survivalsvm, XGBoost-Cox, superpc, and plsRcox) were tuned using 10-fold cross-validation in training and once evaluated in the independent test set. Performance metrics included all time-dependent AUCs, fixed-time ROC AUCs with bootstrap 95% CIs, C-index, various forms of calibration, decision-curve analysis, and simple Kaplan–Meier risk group separation. Results: 193 patients were analyzed (PTS in 64%). High 9-month AUCs were seen in training: GBM (0.992) and RSF (0.982) being the strongest; by 12 months, both remained near constant. Test set performance followed a similar pattern, with RSF again favored (AUC 0.948) and XGBoost/GBM close behind. Calibration was satisfactory, net benefit from decision curves positive, and to a large extent, risk groups were separated as expected. Conclusions: Survival machine-learning models, at least in this dual-center prospective cohort, produced a clinically useful risk of PTS. Hyperuricemia, or any metabolically based signal, is a valuable addition to the “anatomy and care” of DVT. External validation is still required. Full article

(This article belongs to the Collection Artificial Intelligence in Medical Diagnosis and Prognosis)

► Show Figures

Figure 1

11 pages, 792 KB

Open AccessArticle

Quantitative Assessment of Extrinsic Tongue Muscle Stiffness in Obstructive Sleep Apnea Using Shear Wave Elastography

by Hilal Er Ulubaba, Nurullah Dağ, Sevgi Demiröz Taşolar, Rukiye Çiftçi and Hilal Ermiş

Diagnostics 2026, 16(1), 87; https://doi.org/10.3390/diagnostics16010087 (registering DOI) - 26 Dec 2025

Abstract

Background/Objectives: This study aimed to quantitatively and noninvasively evaluate the changes in the Genioglossus (GG) and Geniohyoid (GH) muscles in patients with Obstructive Sleep Apnea (OSA) using ultrasonography (US) and shear wave elastography (SWE). Methods: This prospective study included 94 adults (18–73 years) [...] Read more.

Background/Objectives: This study aimed to quantitatively and noninvasively evaluate the changes in the Genioglossus (GG) and Geniohyoid (GH) muscles in patients with Obstructive Sleep Apnea (OSA) using ultrasonography (US) and shear wave elastography (SWE). Methods: This prospective study included 94 adults (18–73 years) who underwent polysomnography (27 normal; 67 OSA). GG and GH muscle thickness was measured with US, and stiffness with SWE. Participants were grouped as non-OSA (Group 0) and OSA (Group 1). OSA patients were further divided by apnea–hypopnea index (AHI) into mild, moderate, and severe (Groups 1–3), forming four groups including controls. Results: No significant differences were observed in genioglossus or geniohyoid muscle thickness between groups. Shear wave elastography revealed significantly higher stiffness values for both the genioglossus and geniohyoid muscles bilaterally in OSA patients compared with non-OSA individuals (approximately 2.7 m/s vs. 2.4–2.5 m/s, p < 0.01). Geniohyoid muscle stiffness on both sides increased progressively with OSA severity, with significantly higher values in severe compared with mild OSA (p < 0.05). In contrast, genioglossus stiffness did not differ significantly across OSA severity subgroups. Conclusions: In patients with OSA, GH and GG muscle thickness remains unchanged, but their stiffness measured by SWE increases. GH stiffness also rises with increasing disease severity. These results indicate that GG and GH muscle stiffness may serve as useful noninvasive markers for OSA. Full article

(This article belongs to the Special Issue Recent Advances in Diagnosis and Management of Musculoskeletal Disorders)

► Show Figures

Figure 1

17 pages, 438 KB

Open AccessArticle

Prevalence and Interplay of Modifiable and Genetic Determinants of Eustachian Tube Dysfunction Among Saudi Adults: A Nationwide Study

by Mohammad A. Jareebi, Riyadh A. Jahlan, Abdulrahman A. Otaif, Abdulelah A. Otaif, Abdulrahman A. Daghreeri, Mashael S. Mahnashi, Raghad W. Al Nahwe, Yahya A. Maslamani, Ali Y. Mashragi, Abdullah Mawkili, Wedad Mawkili, Faisal Hakami, Sulaiman Ahmed Hussain Darbashi, Majed A. Ryani and Ahmed A. Bahri

Diagnostics 2026, 16(1), 86; https://doi.org/10.3390/diagnostics16010086 - 26 Dec 2025

Abstract

Background/Objectives: Eustachian Tube Dysfunction (ETD) is a prevalent condition affecting middle ear pressure regulation, yet nationwide epidemiological data in Saudi Arabia remain limited. This study aimed to assess the prevalence of ETD and identify its associated factors among Saudi adults using a [...] Read more.

Background/Objectives: Eustachian Tube Dysfunction (ETD) is a prevalent condition affecting middle ear pressure regulation, yet nationwide epidemiological data in Saudi Arabia remain limited. This study aimed to assess the prevalence of ETD and identify its associated factors among Saudi adults using a validated screening tool. Methods: A nationwide cross-sectional study was conducted between June 2024 and March 2025 among 1124 Saudi adults (aged ≥ 18 years) across five geographic regions. ETD was assessed using the validated Arabic version of the seven-item Eustachian Tube Dysfunction Questionnaire (ETDQ-7), with scores ≥ 14.5 indicating dysfunction. Data on demographic, anthropometric, clinical, and lifestyle characteristics were collected via an online questionnaire. Multiple linear regression analysis identified independent predictors of ETDQ-7 scores, with statistical significance set at p < 0.05. Results: The prevalence of ETD was 33.9% (95% CI: 31.1–36.8%), substantially higher than the 7% self-reported rate. Of affected participants, 29.6% had mild-to-moderate ETD and 4.3% had severe dysfunction. Multivariable regression analysis identified four significant independent predictors: higher body mass index (BMI) (β = 0.08; 95% CI: 0.03–0.16; p = 0.049), family history of hearing loss (β = 1.87; 95% CI: 0.90–2.83; p < 0.001), prior bariatric bypass surgery (β = 14.37; 95% CI: 3.33–25.41; p = 0.011), and allergies (β = 3.19; 95% CI: 2.30–4.07; p < 0.001). No significant associations were found with demographic factors, smoking, or other comorbidities. Conclusions: ETD affects approximately one-third of Saudi adults, with significant underdiagnosis. Obesity, genetic predisposition, bariatric surgery, and allergic conditions represent key modifiable and non-modifiable risk factors. These findings support implementing routine ETDQ-7 screening in primary care and targeted interventions for high-risk populations. Full article

(This article belongs to the Special Issue Recent Advances in Hearing and Audiology Science: Diagnosis and Management)

► Show Figures

Figure 1

16 pages, 6481 KB

Open AccessArticle

Biomarker-Based Precision Prediction of Immunotherapy Response in Hepatocellular Carcinoma

by Hsu-Wen Chao, Yi-Mei Joy Lin and Chen-Shiou Wu

Diagnostics 2026, 16(1), 85; https://doi.org/10.3390/diagnostics16010085 (registering DOI) - 26 Dec 2025

Abstract

Background: Hepatocellular carcinoma (HCC) remains a major global health challenge with limited treatment options for advanced disease. Although immune checkpoint inhibitors (ICIs) have shown clinical benefits, response rates remain low, emphasizing the need for reliable biomarkers to guide patient selection. Given the [...] Read more.

Background: Hepatocellular carcinoma (HCC) remains a major global health challenge with limited treatment options for advanced disease. Although immune checkpoint inhibitors (ICIs) have shown clinical benefits, response rates remain low, emphasizing the need for reliable biomarkers to guide patient selection. Given the critical role of metabolic reprogramming in immune modulation, this study aimed to identify a metabolic gene signature predictive of immunotherapy response in HCC. Methods: Three independent transcriptomic datasets (GSE279750, GSE215011, and GSE235863) comprising 35 ICI-treated HCC samples were integrated after quality control and ComBat batch correction. Differentially expressed genes were identified using DESeq2 and limma, followed by integration of the meta-analysis results. Machine learning models, including LASSO regression and random forest algorithms, were applied for feature selection, and a logistic regression model was developed for predictive scoring. Results: A five-gene metabolic signature (PLPPR1, CNTN3, HOXA10, HAGLR, and ENPP3) demonstrated good discriminative ability between responders and non-responders, with consistent performance observed across internal validation analyses. Functional enrichment analysis revealed significant involvement of metabolic pathways, with HOXA10 linked to immune evasion and CNTN3 associated with immune activation. Conclusions: This five-gene signature represents a biologically interpretable biomarker panel with potential utility for immunotherapy response stratification in HCC. The integrative analytical framework provides preliminary evidence supporting its value, warranting further validation in larger, independent clinical cohorts before clinical translation. Full article

(This article belongs to the Special Issue A New Era in Diagnosis: From Biomarkers to Artificial Intelligence)

► Show Figures

Figure 1

27 pages, 3106 KB

Open AccessArticle

An Adaptive Hybrid Metaheuristic Algorithm for Lung Cancer in Pathological Image Segmentation

by Muhammed Faruk Şahin and Ferzat Anka

Diagnostics 2026, 16(1), 84; https://doi.org/10.3390/diagnostics16010084 (registering DOI) - 26 Dec 2025

Abstract

Background/Objectives: Histopathological images are fundamental for the morphological diagnosis and subtyping of lung cancer. However, their high resolution, color diversity, and structural complexity make automated segmentation highly challenging. This study aims to address these challenges by developing a novel hybrid metaheuristic approach for [...] Read more.

Background/Objectives: Histopathological images are fundamental for the morphological diagnosis and subtyping of lung cancer. However, their high resolution, color diversity, and structural complexity make automated segmentation highly challenging. This study aims to address these challenges by developing a novel hybrid metaheuristic approach for multilevel image thresholding to enhance segmentation accuracy and computational efficiency. Methods: An adaptive hybrid metaheuristic algorithm, termed SCSOWOA, is proposed by integrating the Sand Cat Swarm Optimization (SCSO) algorithm with the Whale Optimization Algorithm (WOA). The algorithm combines the exploration capacity of SCSO with the exploitation strength of WOA in a sequential and adaptive manner. The model was evaluated on histopathological images of lung cancer from the LC25000 dataset with threshold levels ranging from 2 to 12, using PSNR, SSIM, and FSIM as performance metrics. Results: The proposed algorithm achieved stable and high-quality segmentation results, with average values of 27.9453 dB in PSNR, 0.8048 in SSIM, and 0.8361 in FSIM. At the threshold level of T = 12, SCSOWOA obtained the highest performance, with SSIM and FSIM scores of 0.9340 and 0.9542, respectively. Furthermore, it demonstrated the lowest average execution time of 1.3221 s, offering up to a 40% improvement in computational efficiency compared with other metaheuristic methods. Conclusions: The SCSOWOA algorithm effectively balances exploration and exploitation processes, providing high-accuracy, low-variance, and computationally efficient segmentation. These findings highlight its potential as a robust and practical solution for AI-assisted histopathological image analysis and lung cancer diagnosis systems. Full article

(This article belongs to the Special Issue Advances in Lung Cancer Diagnosis)

► Show Figures

Graphical abstract

21 pages, 1302 KB

Open AccessArticle

Heart Sound Classification with MFCCs and Wavelet Daubechies Analysis Using Machine Learning Algorithms

by Sebastian Guzman-Alfaro, Karen E. Villagrana-Bañuelos, Manuel A. Soto-Murillo, Jorge Isaac Galván-Tejada, Antonio Baltazar-Raigosa, Angel Garcia-Duran, José María Celaya-Padilla and Andrea Acuña-Correa

Diagnostics 2026, 16(1), 83; https://doi.org/10.3390/diagnostics16010083 - 26 Dec 2025

Abstract

Background/Objectives: Cardiovascular diseases are the leading cause of mortality worldwide according to the World Health Organization (WHO), highlighting the need for accessible tools for early detection. Automated classification systems based on signal processing and machine learning offer a non-invasive alternative to support clinical [...] Read more.

Background/Objectives: Cardiovascular diseases are the leading cause of mortality worldwide according to the World Health Organization (WHO), highlighting the need for accessible tools for early detection. Automated classification systems based on signal processing and machine learning offer a non-invasive alternative to support clinical diagnosis. Methods: This study implements and evaluates machine learning models for distinguishing normal and abnormal heart sounds using a hybrid feature extraction approach. Recordings labeled as normal, murmur, and extrasystolic were obtained from the PASCAL dataset and subsequently binarized into two classes. Multiple numerical datasets were generated through statistical features derived from Mel-Frequency Cepstral Coefficients (MFCCs) and Daubechies wavelet analysis. Each dataset was standardized and used to train four classifiers: support vector machines, logistic regression, random forests, and decision trees. Results: Model performance was assessed using accuracy, precision, recall, specificity, F1-score, and area under curve. All classifiers achieved notable results; however, the support vector machine model trained with 26 MFCCs and Daubechies-4 wavelet coefficients obtained the best performance. Conclusions: These findings demonstrate that the proposed hybrid MFCC–Wavelet framework provides competitive diagnostic accuracy and represents a lightweight, interpretable, and computationally efficient solution for computer-aided auscultation and early cardiovascular screening. Full article

(This article belongs to the Special Issue Artificial Intelligence and Computational Methods in Cardiology 2025)

► Show Figures

Figure 1

14 pages, 817 KB

Open AccessArticle

Optimization of the Contrast Concentration for Low-Tube-Voltage Chest CT: A Randomized Controlled Trial with Low-Concentration Contrast Media

by Jiho Park, Bo Mi Gil, Yun-Hyeon Kim, Gong Yong Jin, Yeon Joo Jeong, Jin Mo Goo, Suyon Chang, Kyongmin Sarah Beck and Jung Im Jung

Diagnostics 2026, 16(1), 82; https://doi.org/10.3390/diagnostics16010082 - 25 Dec 2025

Abstract

Objective: To evaluate the effect of low-concentration contrast media (LCCMs) on 100 kVp conventional chest CT by comparing the proportions of acceptable-quality scans obtained using various CM concentrations with those obtained using the standard protocol. Materials and methods: This prospective, multicenter, [...] Read more.

Objective: To evaluate the effect of low-concentration contrast media (LCCMs) on 100 kVp conventional chest CT by comparing the proportions of acceptable-quality scans obtained using various CM concentrations with those obtained using the standard protocol. Materials and methods: This prospective, multicenter, randomized controlled trial enrolled adult patients with BMI < 30 who underwent enhanced chest CT between December 2023 and September 2024. Participants were randomly assigned to four groups: one control group (120 kVp, 320 mgI/mL) and three test groups using 100 kVp and standard CM (320 mgI/mL) or LCCMs (270 or 240 mgI/mL). Non-inferiority was tested based on the proportion of acceptable-quality scans independently assessed by two readers. Adverse reactions, including injection site pain and localized and systemic heat sensations, were recorded. Results: A total of 371 participants (mean age: 66.0, IQR: 58–73 years) were randomized. 100 kVp chest CT with LCCM groups showed non-inferior image quality to the standard protocol (270 mgI/mL: reader 1—98.33% confidence interval [CI]: −4.95, 9.37 and reader 2–98.33% CI: −7.11, 7.21; 240 mgI/mL: reader 1–98.33% CI: −6.44, 8.71 and reader 2–98.33% CI: −11.47, 5.11; all p < 0.001). Reader 2 reported a lower proportion of acceptable scans in the 320 mgI/mL low-voltage group (p = 0.11), whereas Reader 1 did not find significant inferiority (p = 0.003). Injection site pain, localized heat, and systemic heat were less frequent in LCCM groups, while systemic heat was only statistically significant (p = 0.003). Conclusions: Low-tube-voltage (100 kVp) chest CECT using LCCMs yields non-inferior image quality compared with the standard protocol while using conventional concentration contrast media under 100 kVp resulted in decreased proportion of diagnostically acceptable scans. Full article

(This article belongs to the Section Medical Imaging and Theranostics)

► Show Figures

Graphical abstract

16 pages, 735 KB

Open AccessArticle

Multidimensional Morphology of the Ethmoid Roof and Anterior Ethmoidal Artery: A CT-Based Analysis and Proposal of the Akcan Classification

by Abdullah Belada, Fatih Alper Akcan, Derya Güçlü, Ender Güçlü, İlhan Ünlü, Buğra Subaşı, Mehmet Ali Özel, Ethem İlhan, Derya Cebeci and Mehmet Ali Sungur

Diagnostics 2026, 16(1), 81; https://doi.org/10.3390/diagnostics16010081 - 25 Dec 2025

Abstract

Background/Objectives: Anatomical variation in the ethmoid roof and lateral lamella play an important role in anatomical vulnerability during endoscopic sinus and skull base surgery. However, widely used classifications, including the Keros system, primarily focus on vertical depth and may not fully reflect [...] Read more.

Background/Objectives: Anatomical variation in the ethmoid roof and lateral lamella play an important role in anatomical vulnerability during endoscopic sinus and skull base surgery. However, widely used classifications, including the Keros system, primarily focus on vertical depth and may not fully reflect the complex geometric relationship between the ethmoid roof, lateral lamella, and the anterior ethmoidal artery (AEA). This study aimed to characterize ethmoid roof and lateral lamella anatomy using high-resolution CT and to propose a descriptive radiological framework—the Akcan Classification—that integrates AEA exit patterns with multiple morphometric parameters. Given the complexity of thin skull base structures, interobserver reproducibility of all morphometric parameters was additionally assessed to ensure measurement robustness. Methods: High-resolution paranasal sinus CT scans from 175 adults (350 sides) were retrospectively evaluated. Measurements included ethmoid roof width, lateral lamella depth, anterior–posterior length, lamellar angle, AEA–lamella distance, and sinonasal anatomical variations. Interobserver reliability was quantified using ICCs. AEA morphology was categorized as in-canal (Type 1), partially suspended (Type 2), or fully suspended (Type 3) based on radiological appearance of bony canalization. Appropriate statistical tests were used to compare morphometric features across groups. Results: Suspended AEA configurations demonstrated progressively wider ethmoid roofs, deeper lateral lamellae, steeper lamellar inclination, and shorter AEA–lamella distances (all p < 0.001). Supraorbital ethmoid cells were more frequently observed in Type 3 cases (p < 0.001). Other anatomical variations showed no significant association with ethmoid roof morphology. Interobserver reliability was excellent for all measurements (ICC range 0.87–0.94). Conclusions: The findings suggest that AEA configuration is associated with broader patterns of ethmoid roof and lateral lamella morphology. Rather than serving as a validated predictor of surgical outcomes, the Akcan Classification provides a structured anatomical and radiological descriptor that complements depth-based systems such as the Keros classification. The high reproducibility of measurements supports its potential utility for standardized anatomical assessment and preoperative radiological interpretation, while further studies incorporating surgical correlation are required. Full article

(This article belongs to the Section Medical Imaging and Theranostics)

► Show Figures

Figure 1

Journal Description

Diagnostics

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Topical Collections

Further Information

Guidelines

MDPI Initiatives

Follow MDPI