Diagnostics

19 pages, 38450 KB

Open AccessArticle

Color Normalization in Breast Cancer Immunohistochemistry Images Based on Sparse Stain Separation and Self-Sparse Fuzzy Clustering

by Attasuntorn Traisuwan, Somchai Limsiroratana, Pornchai Phukpattaranont, Phiraphat Sutthimat and Pichaya Tandayya

Diagnostics 2025, 15(18), 2316; https://doi.org/10.3390/diagnostics15182316 - 12 Sep 2025

Abstract

Background and Objective: The color normalization of breast cancer immunohistochemistry (IHC)-stained images helps change the color distribution of undesirable IHC-stained images to be more interpretable for the pathologists. This will affect the Allred score that the pathologists use to estimate the drug [...] Read more.

Background and Objective: The color normalization of breast cancer immunohistochemistry (IHC)-stained images helps change the color distribution of undesirable IHC-stained images to be more interpretable for the pathologists. This will affect the Allred score that the pathologists use to estimate the drug quantity for treating breast cancer patients. Methods: A new color normalization technique based on sparse stain separation and self-sparse fuzzy clustering is proposed. Results: The quaternion structural similarity was used to measure the quality of the normalization algorithm. Our technique has a structural similarity score lower than other techniques, and the color distribution similarity is closer to the target. We applied automated and unsupervised nuclei classification with Automatic Color Deconvolution (ACD) to test the color features extracted from normalized images. Conclusions: The classification result from our unsupervised nuclei classification with ACD is similar to other normalization methods, but it offers an easier perception to the pathologists. Full article

(This article belongs to the Special Issue Medical Images Segmentation and Diagnosis)

► Show Figures

Figure 1

14 pages, 851 KB

Open AccessArticle

Performance of a Vision-Language Model in Detecting Common Dental Conditions on Panoramic Radiographs Using Different Tooth Numbering Systems

by Zekai Liu, Qi Yong H. Ai, Andy Wai Kan Yeung, Ray Tanaka, Andrew Nalley and Kuo Feng Hung

Diagnostics 2025, 15(18), 2315; https://doi.org/10.3390/diagnostics15182315 - 12 Sep 2025

Abstract

Objectives: The aim of this study was to evaluate the performance of GPT-4o in identifying nine common dental conditions on panoramic radiographs, both overall and at specific tooth sites, and to assess whether the use of different tooth numbering systems (FDI and [...] Read more.

Objectives: The aim of this study was to evaluate the performance of GPT-4o in identifying nine common dental conditions on panoramic radiographs, both overall and at specific tooth sites, and to assess whether the use of different tooth numbering systems (FDI and Universal) in prompts would affect its diagnostic accuracy. Methods: Fifty panoramic radiographs exhibiting various common dental conditions including missing teeth, impacted teeth, caries, endodontically treated teeth, teeth with restorations, periapical lesions, periodontal bone loss, tooth fractures, cracks, retained roots, dental implants, osteolytic lesions, and osteosclerosis were included. Each image was evaluated twice by GPT-4o in May 2025, using structured prompts based on either the FDI or Universal tooth numbering system, to identify the presence of these conditions at specific tooth sites or regions. GPT-4o responses were compared to a consensus reference standard established by an oral-maxillofacial radiology team. GPT-4o’s performance was evaluated using balanced accuracy, sensitivity, specificity, and F1 score both at the patient and tooth levels. Results: A total of 100 GPT-4o responses were generated. At the patient level, balanced accuracy ranged from 46.25% to 98.83% (FDI) and 49.75% to 92.86% (Universal), with the highest accuracies for dental implants (92.86–98.83%). F1-scores and sensitivities were highest for implants, missing, and impacted teeth, but zero for caries, periapical lesions, and fractures. Specificity was generally high across conditions. Notable discrepancies were observed between patient- and tooth-level performance, especially for implants and restorations. GPT-4o’s performance was similar between using the two numbering systems. Conclusions: GPT-4o demonstrated superior performance in detecting dental implants and treated or restored teeth but inferior performance for caries, periapical lesions, and fractures. Diagnostic accuracy was higher at the patient level than at the tooth level, with similar performances for both numbering systems. Future studies with larger, more diverse datasets and multiple models are needed. Full article

(This article belongs to the Special Issue Advances in Head and Neck and Oral Maxillofacial Radiology)

► Show Figures

Figure 1

22 pages, 855 KB

Open AccessSystematic Review

Prevalence of Tuberculosis in Central Asia and Southern Caucasus: A Systematic Literature Review

by Malika Idayat, Elena von der Lippe, Nailya Kozhekenova, Oyunzul Amartsengel, Kamila Akhmetova, Ainash Oshibayeva, Zhansaya Nurgaliyeva and Natalya Glushkova

Diagnostics 2025, 15(18), 2314; https://doi.org/10.3390/diagnostics15182314 - 12 Sep 2025

Abstract

Background: In 2023, tuberculosis (TB) caused 1.25 million deaths globally, remaining a leading infectious killer. Central Asia and Southern Caucasus face high TB burdens, particularly Mongolia. This review synthesizes TB prevalence data and diagnostic capabilities in these regions to support public health [...] Read more.

Background: In 2023, tuberculosis (TB) caused 1.25 million deaths globally, remaining a leading infectious killer. Central Asia and Southern Caucasus face high TB burdens, particularly Mongolia. This review synthesizes TB prevalence data and diagnostic capabilities in these regions to support public health strategies. Methods: This systematic review aimed to synthesize current data on TB prevalence in Central Asia, Southern Caucasus, and Mongolia to support public health strategies and research priorities. A comprehensive search of PubMed and Google Scholar was conducted for English-language articles published up to 2023. Studies were assessed using a modified Newcastle–Ottawa Scale. Nine studies met the inclusion criteria, covering Kazakhstan, Kyrgyzstan, Uzbekistan, Tajikistan, Turkmenistan, Mongolia, Georgia, Armenia, and Azerbaijan. Results: TB incidence ranged from 67 per 100,000 in Kazakhstan to 190 per 100,000 in Kyrgyzstan, with the highest prevalence of 68.5% in Mongolia. TB affected men more frequently (65.3%), and the key risk factors included HIV (30.5%), comorbidities, and undernutrition. Diagnostic performance varied significantly (microscopy sensitivity, 45–65%; GeneXpert MTB/RIF, 89–96% sensitivity and 98% specificity for rifampicin resistance). Diagnostic turnaround times ranged from hours (molecular) to weeks (conventional). Only 58% of TB facilities had GeneXpert technology, with urban–rural disparities in diagnostic access. Drug-resistant TB imposed a significant economic burden, with treatment costs ranging from USD 106 to USD 3125. Conclusions: Strengthening surveillance, improving data collection, and conducting longitudinal studies are essential for designing effective TB control strategies in these regions. Significant diagnostic gaps persist across these regions, especially with regard to drug-resistant strains. Point-of-care molecular diagnostics, improved algorithms, and expanded laboratory training show promise. Future research should focus on rapid biomarker-based diagnostics, field-deployable technologies for settings with limited resources, and AI integration to enhance diagnostic accuracy and efficiency. Full article

(This article belongs to the Special Issue Recent Advances in Epidemiological Diagnostics: Detecting and Controlling Infectious Diseases)

► Show Figures

Figure 1

16 pages, 1547 KB

Open AccessArticle

Long-Term Retrospective Analysis of Parvovirus B19 Infections in Blood Donors (2012–2024): Significant Increase in Prevalence Following the SARS-CoV-2 Pandemic

by Michaela Oeller, Orkan Kartal, Iuliia Trifonova, Nina Held, Alexandra Domnica Hoeggerl, Heidrun Neureiter, Wanda Lauth, Christoph Grabmer, Eva Rohde and Sandra Laner-Plamberger

Diagnostics 2025, 15(18), 2313; https://doi.org/10.3390/diagnostics15182313 - 11 Sep 2025

Abstract

Background/Objectives: Parvovirus B19 (B19V) is a non-enveloped single-stranded DNA virus transmissible by blood transfusion, with potentially severe outcomes in immunocompromised and pregnant recipients. In this study, we investigated the B19V prevalence in 441,084 blood donations from Salzburg, Austria, collected between 2012 and [...] Read more.

Background/Objectives: Parvovirus B19 (B19V) is a non-enveloped single-stranded DNA virus transmissible by blood transfusion, with potentially severe outcomes in immunocompromised and pregnant recipients. In this study, we investigated the B19V prevalence in 441,084 blood donations from Salzburg, Austria, collected between 2012 and 2024, focusing on changes in epidemiological dynamics before, during, and after the SARS-CoV-2 pandemic. Additionally, the B19VB19V persistence and its implications for deferral policies were assessed. Methods: Donor samples were screened for B19VB19V DNA by qPCR (2012–2024) and for SARS-CoV-2 total anti-N antibodies (2020–2024). B19VB19V prevalence rates, cycle threshold (Ct) values, and seasonal distribution were compared between pre-pandemic, pandemic, and post-pandemic phases. Follow-up testing of initially B19VB19V-positive donors was performed after a 2-year deferral period. Results: The B19VB19V positivity rate of 0.13% (2012–2019) significantly decreased to 0.02% during the SARS-CoV-2 pandemic (2020–2022). A substantial increase occurred post-pandemic, with prevalence reaching 1.47% in 2024. Significant lower Ct values were observed in the post-pandemic phase, indicating higher viral loads. Additionally, younger donors (aged 18–45 years) showed significantly lower Ct values. After a 2-year deferral, 39% of re-tested donors remained B19VB19V DNA-positive. Conclusions: B19VB19V circulation increased substantially after the SARS-CoV-2 pandemic. Our observation is consistent with international reports and is likely due to an ‘immunity debt’ that has been accumulated due to pandemic-related public health interventions. Targeted B19VB19V screening and strict deferral strategies may be warranted particularly during outbreak periods to protect high-risk transfusion recipients. Full article

(This article belongs to the Special Issue Recent Advances in Epidemiological Diagnostics: Detecting and Controlling Infectious Diseases)

8 pages, 850 KB

Open AccessInteresting Images

A Cystic-like Lesion of Uncertain Origin—A Discussion on Cemento-Osseous Dysplasia and Traumatic Bone Cysts

by Kamil Nelke, Maciej Karpiński, Michał Scharoch, Maciej Janeczek, Agata Małyszek, Evagelos Kalfarentzos, Efthymios Mavrakos, Piotr Kuropka, Christos Perisanidis and Maciej Dobrzyński

Diagnostics 2025, 15(18), 2312; https://doi.org/10.3390/diagnostics15182312 - 11 Sep 2025

Abstract

Mandible cemento-osseous dysplasia (COD) can be found mostly associated with dental roots and tooth-bearing anatomical structures. A variety of odontogenic cysts and tumors might have similar appearances. A lesion in the jaw bone not associated with dental roots with a cyst-like appearance might [...] Read more.

Mandible cemento-osseous dysplasia (COD) can be found mostly associated with dental roots and tooth-bearing anatomical structures. A variety of odontogenic cysts and tumors might have similar appearances. A lesion in the jaw bone not associated with dental roots with a cyst-like appearance might suggest a non-odontogenic lesion, an empty bone cavity, an osseous, fibrous, or fibro-osseous lesion, or a traumatic bone cyst (TBC). A radiolucent irregular bone cavity without clear borders always requires improved diagnostics in cone-beam computed tomography (CBCT) as well as a revision and a biopsy in some cases. When there is some bone swelling and asymmetry on radiological evaluation, followed by extra-cortical spread, and the lesion has irregular borders with thickening or atypical calcifications, a biopsy should be performed. COD and TBCs can be found mostly associated with dental roots, but sometimes they are not associated with tooth-bearing jaw structures and might cause some diagnostic problems, especially if they resemble an empty radiolucent cystic-like lesion in an atypical location. Regardless of the type of lesion, a bone revision and a biopsy are important. When a sufficient amount of a sample is removed and evaluated, this can greatly improve the final diagnosis. The authors present an interesting case of a lesion accidentally found in a routine panoramic radiograph used for screening before scheduled orthodontic treatment. Full article

(This article belongs to the Collection Interesting Images)

► Show Figures

Figure 1

15 pages, 4386 KB

Open AccessArticle

Microstructural Analysis of Whole-Brain Changes Increases the Detection of Pediatric Focal Cortical Dysplasia

by Xinyi Yang, Shuang Ding, Song Peng, Wei Tang, Yali Gao, Zhongxin Huang and Jinhua Cai

Diagnostics 2025, 15(18), 2311; https://doi.org/10.3390/diagnostics15182311 - 11 Sep 2025

Abstract

Purpose: Focal cortical dysplasia (FCD) is a common developmental malformation disease of the cerebral cortex. Although mounting evidence has suggested that FCD lesions have variable locations and topographies throughout the cortex, few studies have explored consistencies in structural connectivity among different lesion [...] Read more.

Purpose: Focal cortical dysplasia (FCD) is a common developmental malformation disease of the cerebral cortex. Although mounting evidence has suggested that FCD lesions have variable locations and topographies throughout the cortex, few studies have explored consistencies in structural connectivity among different lesion types. In this study, we analyzed microscopic structural changes via lesion analysis and explored structural changes in nonlesion regions across the brain. Methods: Diffusion tensor imaging (DTI) and magnetization transfer imaging were used to compare FCD lesions and contralateral normal appearing gray/white matter (cNAG/WM). Voxel-based morphometry was calculated for 28 children with FCD and 34 sex- and age-matched healthy participants. DTI indices of the FCD and healthy control groups were analyzed via the tract-based spatial statistic method to evaluate the microstructure abnormalities of WM fiber tracts in individuals with FCD. Results: In terms of FCD lesions, compared with those of the cNAG, the fractional anisotropy (FA) values were decreased, and the mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD) values were increased; the magnetization transfer ratios were also decreased. In terms of whole-brain changes due to FCD, compared with the healthy control group, the FCD group showed a decrease in the volume of the right hippocampus and left anterior cingulate cortex. FCD patients had lower FA values, higher MD values, lower AD values, and mainly increased RD values in relation to WM microstructure. Conclusions: Microstructural abnormalities outside lesion regions may be related to injury to the epileptic network, and the identification of such abnormalities may complement diagnoses of FCD in pediatric patients. Full article

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

► Show Figures

Figure 1

19 pages, 3431 KB

Open AccessArticle

Computed Tomography Radiomics and Machine Learning for Prediction of Histology-Based Hepatic Steatosis Scores

by Winston T. Chu, Hui Wang, Marcelo A. Castro, Venkatesh Mani, C. Paul Morris, Thomas C. Friedrich, David H. O’Connor, Courtney L. Finch, Ji Hyun Lee, Philip J. Sayre, Gabriella Worwa, Anya Crane, Jens H. Kuhn, Ian Crozier, Jeffrey Solomon and Claudia Calcagno

Diagnostics 2025, 15(18), 2310; https://doi.org/10.3390/diagnostics15182310 - 11 Sep 2025

Abstract

Background/Objective: Computed tomography (CT) can be used to non-invasively assess the health of the liver; however, radiologist evaluation and simple thresholding alone are insufficient for diagnosis of hepatic steatosis, necessitating biopsies. This study explored CT radiomics and machine learning to enable non-invasive, objective, [...] Read more.

Background/Objective: Computed tomography (CT) can be used to non-invasively assess the health of the liver; however, radiologist evaluation and simple thresholding alone are insufficient for diagnosis of hepatic steatosis, necessitating biopsies. This study explored CT radiomics and machine learning to enable non-invasive, objective, and quantitative prediction of steatosis severity across the macaque liver. Methods: In this retrospective study, CT images of 42 crab-eating macaques (age [yr] = 6.1 ± 1.7; sex [male/female] = 26/16) with varying degrees of hepatic steatosis were analyzed, and the results were compared to histology-based steatosis scores of livers from the same animals. After extracting radiomic features, a thorough array of statistical analyses, feature selection techniques, and machine learning models were applied to identify a distinct radiomic signature of histologically defined hepatic steatosis. Results: We identified 12 radiomic features that correlated with steatosis scores, and hierarchical clustering based on radiomic attributes alone revealed clusters roughly aligning with steatosis severity groups. The k-nearest neighbors model architecture best predicted histopathologic steatosis scores in both classification and regression tasks (area under the receiver operating characteristic curve [AUC ROC] = 0.89 ± 0.09; root-mean-square error [RMSE] = 0.60 ± 0.10). Feature analyses identified seven key radiomic features (six first-order features and one gray-level co-occurrence matrix feature) that were most important when predicting steatosis. Conclusions: We identified a CT radiomic signature of steatosis and demonstrated that histology-based steatosis scores can be predicted non-invasively and objectively using machine learning and CT radiomics as a potential alternative to invasive core biopsies. Given the strong similarities in liver structure, liver function, and hepatic steatosis pathophysiology between macaques and humans, these findings have the potential to translate to humans. Full article

(This article belongs to the Special Issue Artificial Intelligence-Driven Radiomics in Medical Diagnosis)

► Show Figures

Figure 1

18 pages, 760 KB

Open AccessArticle

Antineutrophil Cytoplasmic Autoantibodies Specific to Bactericidal/Permeability-Increasing Protein: A Cross-Road Between Prolonged Gram-Negative Bacterial Infections and Ulcerative Colitis/Primary Sclerosing Cholangitis

by Dragana Jovanovic, Rada Miskovic, Aleksandra Plavsic, Sara Radovic, Ljudmila Nagorni-Obradovic, Dragan Popovic, Milos M. Nikolic and Branka Bonaci-Nikolic

Diagnostics 2025, 15(18), 2309; https://doi.org/10.3390/diagnostics15182309 - 11 Sep 2025

Abstract

Background/Objectives: Binding of bactericidal/permeability-increasing (BPI) protein to Gram-negative (GN) bacteria plays a major role in bacterial elimination. The relationship between BPI-antineutrophil cytoplasmic autoantibodies (ANCA), persistent infections and immunoinflammatory diseases has not been elucidated. Methods: In total, 193 ANCA-positive patients detected by [...] Read more.

Background/Objectives: Binding of bactericidal/permeability-increasing (BPI) protein to Gram-negative (GN) bacteria plays a major role in bacterial elimination. The relationship between BPI-antineutrophil cytoplasmic autoantibodies (ANCA), persistent infections and immunoinflammatory diseases has not been elucidated. Methods: In total, 193 ANCA-positive patients detected by IIF with ANCA-associated vasculitides (AAV, n-40), connective tissue diseases (CTD, n-28), drug-induced vasculitides (DIV, n-17), ulcerative colitis (UC, n-24), UC with primary sclerosing cholangitis (UC/PSC, n-14), Crohn’s disease (CD, n-10), autoimmune hepatitis (AIH, n-19) and chronic infections (n-41) were tested using the BPI-ANCA quantitative and semiquantitative ELISA (ANCA-profile: BPI, proteinase 3, myeloperoxidase, elastase, cathepsin G, lactoferrin). BPI-ANCA were analyzed in 52 healthy persons. Results: A total of 46/193 (23.8%) patients had BPI-ANCA positivity. BPI-ANCA were more frequently present in patients with prolonged GN bacterial infections and inflammatory bowel diseases than in AAV, DIV, AIH, CTD and healthy controls (p < 0.001). UC/PSC patients more frequently had BPI-ANCA than UC and CD patients (p < 0.001). GN bacterial infections more frequently had BPI-ANCA than Gram-positive bacterial infections (p < 0.001). Infections caused by Pseudomonas aeruginosa and Mycobacterium tuberculosis had monospecific BPI-ANCA (sensitivity 79% and 71%, respectively). UC/PSC and chronic GN bacterial infections caused by Klebsiella pneumoniae, Proteus mirabilis, or Escherichia coli had multispecific BPI-ANCA (sensitivity 64% and 100%, respectively). Odds ratio analysis showed that patients with IBD who were positive for multispecific BPI-ANCA had a 13.5-fold increased risk of UC/PSC (95% CI 2.98–61.18). Conclusions: Monospecific BPI-ANCA may be a valuable biomarker for persistent Pseudomonas aeruginosa and Mycobacterium tuberculosis infections. In contrast, multispecific BPI-ANCA are associated with UC/PSC and persistent infections caused by intestinal Gram-negative bacteria. Suppression of antimicrobial function by multispecific BPI-ANCA could impair the elimination of Gram-negative bacteria, sustaining the immunoinflammation. Dysregulated antimicrobial response might be the target of immunomodulatory therapy in the initial phase of BPI-ANCA-positive UC/PSC. Full article

(This article belongs to the Special Issue Advances in the Diagnosis of Infectious Diseases and Microorganisms: 2nd Edition)

► Show Figures

Figure 1

16 pages, 957 KB

Open AccessReview

The Rise of AI-Assisted Diagnosis: Will Pathologists Be Partners or Bystanders?

by Riyad El-Khoury and Ghazi Zaatari

Diagnostics 2025, 15(18), 2308; https://doi.org/10.3390/diagnostics15182308 - 11 Sep 2025

Abstract

Over 150 years, pathology has transformed remarkably, from the humble beginnings of microscopic tissue examination to today’s revolutionary advancements in digital pathology and artificial intelligence (AI) applications. This review briefly retraces the evolution of microscopes and highlights breakthroughs in complementary tools and techniques [...] Read more.

Over 150 years, pathology has transformed remarkably, from the humble beginnings of microscopic tissue examination to today’s revolutionary advancements in digital pathology and artificial intelligence (AI) applications. This review briefly retraces the evolution of microscopes and highlights breakthroughs in complementary tools and techniques that laid the foundation for modern surgical pathology, recently expanded into a new dimension with digital pathology. Digital pathology marked a pivotal turning point by addressing the longstanding limitations of conventional microscopy, paving the way for AI integration. AI now revolutionizes pathology workflows, offering unprecedented opportunities for automated diagnostics, enhanced precision, accelerated research, and advanced medical education. Despite widespread consensus on AI as complementary to pathologists, rare studies critically explore the feasibility of a fully autonomous, pathologist-independent diagnostic workflow. Given the rapid advancement of AI, it is timely to examine whether mature AI systems might realistically achieve diagnostic autonomy. Thus, this review uniquely addresses this gap by evaluating the feasibility, limitations, and implications of a disruptive, pathologist-free diagnostic model. This exploration raises critical questions about the evolving role of pathologists in an era increasingly defined by automation. Can pathologists adapt to emerging trends, maintain their central role in patient care, and leverage AI effectively, or will their traditional roles inevitably diminish? Could the continued advancement of AI eventually prompt a return of pathologists to their initial mid-19th century role as scientist scholars, removed from frontline diagnostics? Ultimately, we assess whether AI can independently sustain diagnostic accuracy and decision making without pathologist oversight. Full article

(This article belongs to the Section Machine Learning and Artificial Intelligence in Diagnostics)

► Show Figures

Figure 1

9 pages, 201 KB

Open AccessCommunication

Criteria for Routine Laboratory Blood Tests in Patients Hospitalized in Cardiology Departments

by Zvi Shimoni, Fadi Hin and Paul Froom

Diagnostics 2025, 15(18), 2307; https://doi.org/10.3390/diagnostics15182307 - 11 Sep 2025

Abstract

The proportion of laboratory tests ordered in cardiology departments without clinical utility is unclear. The objective of this study was to determine if criteria limiting testing can safely reduce admission and follow-up testing. We reviewed the charts of 471 consecutive patients admitted to [...] Read more.

The proportion of laboratory tests ordered in cardiology departments without clinical utility is unclear. The objective of this study was to determine if criteria limiting testing can safely reduce admission and follow-up testing. We reviewed the charts of 471 consecutive patients admitted to the cardiology department at a regional hospital from January 2019 to June 2019. We prospectively set appropriate criteria for routine admission and follow-up testing. Commonly ordered tests and parameters considered not to be indicated either on admission or on follow-up included C-reactive protein, liver function tests, lactate dehydrogenase, creatine phosphokinase, calcium, blood urea nitrogen, uric acid, cholesterol, Hemoglobin A1c, and prothrombin times (except for patients treated with warfarin). Admission tests considered appropriate included electrolytes, glucose, creatinine, and complete blood counts. Follow-up testing was indicated only if test results were outside the reference ranges. Troponin tests were only indicated if needed to determine the need for a coronary angiogram. The outcome variables were the proportion of indicated tests and whether tests outside the criteria led to changes in acute care that positively affected the patient’s hospital care. In the 471 patients, there were 18,061 tests ordered (not including troponin), and 14,427 (79.9%) were not indicated; this led to 46 (0.3%) changes in medical care, which did not affect the patients’ clinical course. There were 47.8% (364/761) troponin tests that were not indicated and did not change patient care. Our study suggests that interventions in cardiology departments such as ours could safely reduce troponin testing by nearly 50% and other laboratory tests by around 80%. These results need to be confirmed in other settings and in interventional studies. Full article

(This article belongs to the Section Clinical Laboratory Medicine)

11 pages, 717 KB

Open AccessArticle

[¹²³I]-Meta-Iodobenzylguanidine Scintigraphy in Sarcoidosis: Exploring Cardiac Autonomic Dysfunction in Patients with Unexplained Cardiac Symptoms

by Lisette R. M. Raasing, Marjolein Drent, Ruth G. M. Keijsers, Andor F. van den Hoven, Marco C. Post, Jan C. Grutters and Marcel Veltkamp

Diagnostics 2025, 15(18), 2306; https://doi.org/10.3390/diagnostics15182306 - 11 Sep 2025

Abstract

Background/Objectives: Sarcoidosis is a systemic inflammatory disease that can cause cardiac autonomic dysfunction (SCAD), often underrecognized despite its clinical importance. While [¹⁸F]fluoro-2-deoxy-D-glucose positron emission tomography/computed tomography ([¹⁸F]FDG PET/CT) and cardiac magnetic resonance imaging (CMR) assess cardiac involvement, [ [...] Read more.

Background/Objectives: Sarcoidosis is a systemic inflammatory disease that can cause cardiac autonomic dysfunction (SCAD), often underrecognized despite its clinical importance. While [¹⁸F]fluoro-2-deoxy-D-glucose positron emission tomography/computed tomography ([¹⁸F]FDG PET/CT) and cardiac magnetic resonance imaging (CMR) assess cardiac involvement, [¹²³I]-meta-iodinebenzylguanidine ([¹²³I]MIBG) scintigraphy evaluates cardiac sympathetic innervation, offering complementary insights to potentially improve SCAD detection and management. This retrospective study explores the role of [¹²³I] MIBG scintigraphy in detecting SCAD among patients with unexplained cardiac symptoms. It focuses on its potential to provide complementary diagnostic information in patients where established imaging techniques, such as [¹⁸F]FDG PET/CT and CMR, fail to detect cardiac sarcoidosis. Methods: Sarcoidosis patients referred to the St. Antonius Hospital (2017–2024) who underwent [¹²³I]MIBG scintigraphy were included. Collected data encompassed demographics, SCAD symptoms, cardiac imaging findings, and carvedilol treatment outcomes. [¹²³I] MIBG abnormalities were defined as a heart-to-mediastinal ratio ≤1.6 or a washout rate ≥20%. Results: Among the final cohort of 40 sarcoidosis patients with unexplained cardiac symptoms and normal [¹⁸F]FDG PET/CT and CMR findings, 19 patients (48%) showed abnormal [¹²³I] MIBG scintigraphy results suggestive of SCAD. No significant differences were observed in clinical characteristics between patients with normal and abnormal [¹²³I]MIBG findings. Of the 16 patients treated with carvedilol, 88% reported symptom improvement, although 50% experienced side effects. Conclusions: [¹²³I]MIBG scintigraphy revealed abnormalities in a substantial number of sarcoidosis patients with unexplained cardiac symptoms despite normal [¹⁸F]FDG PET/CT and CMR. These findings indicate a potential role for [¹²³I]MIBG in detecting SCAD, but prospective studies are needed to confirm their clinical significance. Full article

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

► Show Figures

Figure 1

16 pages, 501 KB

Open AccessReview

Radiopharmaceuticals in Malignant Melanoma: A Comprehensive Review of Diagnostic, Therapeutic, and Immune-Related Applications by PET/CT, SPECT/CT, and PET/MRI

by Irina Pirsan and Doina Piciu

Diagnostics 2025, 15(18), 2305; https://doi.org/10.3390/diagnostics15182305 - 11 Sep 2025

Abstract

Background: Malignant melanoma remains an oncological challenge, with advanced-stage five-year survival rates under 20%. Precise molecular imaging has become indispensable for accurate staging, selection of targeted or immunotherapies, treatment response assessment, and early detection of immune-related adverse events. This review examines the roles [...] Read more.

Background: Malignant melanoma remains an oncological challenge, with advanced-stage five-year survival rates under 20%. Precise molecular imaging has become indispensable for accurate staging, selection of targeted or immunotherapies, treatment response assessment, and early detection of immune-related adverse events. This review examines the roles of PET/CT, PET/MRI, and SPECT/CT radiopharmaceuticals in melanoma management and highlights novel tracers and theranostic strategies poised to enhance precision nuclear medicine in this disease. Methods: We performed a review of English-language literature from January 2000 through June 2025, querying PubMed, Scopus, and clinical-trial registries for original research articles, meta-analyses, clinical guidelines, and illustrative case reports. Eligible studies investigated PET/CT, PET/MRI, or SPECT/CT applications in melanoma diagnosis, nodal and distant staging, therapy monitoring, irAE (immune-related adverse events) detection, and the development of emerging radiotracers or theranostic radiopharmaceutical pairs. Results:¹⁸F-FDG PET/CT demonstrated a high detection rate for distant metastases, outperforming conventional CT and MRI in advanced disease, despite limited resolution for infracentimetric nodal deposits. PET/MRI offers comparable diagnostic accuracy with superior soft-tissue contrast and improved brain lesion detection, while SPECT/CT enhanced sentinel lymph node localization prior to surgical biopsy. Also, FDG PET/CT identified visceral irAEs with great sensitivities, revealing asymptomatic toxicities in up to one-third of patients. Emerging radiotracers targeting melanin, fibroblast activation protein, PD-1 (programmed cell death protein 1)/PD-L1 (programmed cell death-ligand 1), and CD8⁺ T cells have demonstrated enhanced tumor specificity and are on their way to forming novel theranostic pairs. Conclusions: While ¹⁸F-FDG PET/CT remains the cornerstone of melanoma imaging, complementary advantages of PET/MRI and SPECT/CT imaging refine melanoma management. The advent of highly specific radiotracers and integrated theranostic approaches heralds a new era of tailored nuclear-medicine strategies, promising improved patient stratification, therapy guidance, and clinical outcomes in melanoma. Full article

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

► Show Figures

Figure 1

24 pages, 5448 KB

Open AccessArticle

GlioSurvQNet: A DuelContextAttn DQN Framework for Brain Tumor Prognosis with Metaheuristic Optimization

by M. Renugadevi, Venkateswarlu Gonuguntla, Ihssan S. Masad, G. Venkat Babu and K. Narasimhan

Diagnostics 2025, 15(18), 2304; https://doi.org/10.3390/diagnostics15182304 - 11 Sep 2025

Abstract

Background/Objectives: Accurate classification of brain tumors and reliable prediction of patient survival are essential in neuro-oncology, guiding clinical decisions and enabling precision treatment planning. However, conventional machine learning and deep learning methods often struggle with challenges such as data scarcity, class imbalance, limited [...] Read more.

Background/Objectives: Accurate classification of brain tumors and reliable prediction of patient survival are essential in neuro-oncology, guiding clinical decisions and enabling precision treatment planning. However, conventional machine learning and deep learning methods often struggle with challenges such as data scarcity, class imbalance, limited model interpretability, and poor generalization across diverse clinical settings. This study presents GlioSurvQNet, a novel reinforcement learning-based framework designed to address these limitations for both glioma grading and survival prediction. Methods: GlioSurvQNet is built upon a DuelContextAttn Deep Q-Network (DQN) architecture, tailored for binary classification of low-grade vs. high-grade gliomas and multi-class survival prediction (short-, medium-, and long-term categories). Radiomics features were extracted from multimodal MRI scans, including FLAIR, T1CE, and T2 sequences. Feature optimization was performed using a hybrid ensemble of metaheuristic algorithms, including Harris Hawks Optimization (HHO), Modified Gorilla Troops Optimization (mGTO), and Zebra Optimization Algorithm (ZOA). Subsequently, SHAP-based feature selection was applied to enhance model interpretability and robustness. Results: The classification module achieved the highest accuracy of 99.27% using the FLAIR + T1CE modality pair, while the survival prediction model attained an accuracy of 93.82% with the FLAIR + T2 + T1CE fusion. Comparative evaluations against established machine learning and deep learning models demonstrated that GlioSurvQNet consistently outperformed existing approaches in both tasks. Conclusions: GlioSurvQNet offers a powerful and interpretable AI-driven solution for brain tumor analysis. Its high accuracy and robustness make it a promising tool for clinical decision support in glioma diagnosis and prognosis. Full article

(This article belongs to the Section Machine Learning and Artificial Intelligence in Diagnostics)

► Show Figures

Figure 1

11 pages, 689 KB

Open AccessArticle

Fetal Cardiovascular Profile Score (CVPs) in Fetal Anemia, Using Fetal Hemoglobin Bart’s Disease at Mid-Pregnancy as a Study Model

by Panisa Hantrakun, Kasemsri Srisupundit and Theera Tongsong

Diagnostics 2025, 15(18), 2303; https://doi.org/10.3390/diagnostics15182303 - 11 Sep 2025

Abstract

Objectives: To evaluate the diagnostic performance of CVPs in predicting fetal Hb Bart’s disease among pregnancies at risk and to study hemodynamic changes based on CVP components in response to fetal anemia. Methods: The database was assessed to retrieve the ultrasound records of [...] Read more.

Objectives: To evaluate the diagnostic performance of CVPs in predicting fetal Hb Bart’s disease among pregnancies at risk and to study hemodynamic changes based on CVP components in response to fetal anemia. Methods: The database was assessed to retrieve the ultrasound records of fetuses at risk of Hb Bart’s disease at 17–22 weeks and the relevant files including complete video sets of fetal echocardiography. The five components of CVPs of each case were blindly assigned. The definitive diagnosis of fetal Hb Bart’s disease was based on cordocentesis or neonatal blood analysis. Results: Among 378 pregnancies at risk that were recruited into the study, there were 76 (20.1%) affected fetuses and 302 (79.9%) unaffected fetuses. Using a cut-off score of <9, CVPs had a sensitivity of 92.1% and specificity of 97.4% in predicting affected fetuses. However, the effectiveness was not much superior to cardio-thoracic area ratio (CTAR) alone (area under curve; AUC: 0.983 vs. 0.954). Of all parameters, CTAR provided the best diagnostic performance. The combination of CTAR and assessment of hydropic sign provided the best diagnostic values, comparable with full CVPs (AUC 0.982 vs. 0.983). The affected fetuses cope well with anemia by physically increasing in cardiac size and functionally increasing in Tei index with minimally reduced shortening fraction, without compromising arterial and venous Doppler indices. Conclusions: CVPs are highly effective in predicting affected fetuses among pregnancies at risk of fetal Hb Bart’s disease. Nevertheless, only two components (CTAR and hydropic sign) are adequate to yield the best diagnostic performance. Full article

(This article belongs to the Special Issue Advances in Fetal Cardiology)

► Show Figures

Figure 1

4 pages, 526 KB

Open AccessInteresting Images

Surgical Management of a Ruptured Giant Left Main Coronary Artery Aneurysm Presenting with Cardiac Tamponade

by Dmitriy Shumakov, Dmitriy Zybin, Elena Stepanova, Siarhei Dabravolski, Elena Sigaleva, Ekaterina Silina, Victor Stupin and Mikhail Popov

Diagnostics 2025, 15(18), 2302; https://doi.org/10.3390/diagnostics15182302 - 10 Sep 2025

Abstract

Coronary artery aneurysms (CAAs) are an uncommon finding, and their rupture is an exceedingly rare and life-threatening complication. Giant aneurysms of the left main coronary artery (LMCA) pose a significant diagnostic and therapeutic challenge. We describe the case of a 62-year-old male who [...] Read more.

Coronary artery aneurysms (CAAs) are an uncommon finding, and their rupture is an exceedingly rare and life-threatening complication. Giant aneurysms of the left main coronary artery (LMCA) pose a significant diagnostic and therapeutic challenge. We describe the case of a 62-year-old male who presented with acute coronary syndrome and was subsequently diagnosed with a ruptured giant LMCA aneurysm causing cardiac tamponade and multi-organ dysfunction. The initial diagnosis was suggested by coronary angiography and confirmed with contrast-enhanced multidetector computed tomography (MDCT) and echocardiography. The patient underwent emergency surgery consisting of aneurysm excision, thrombectomy, ligation of the LMCA ostium and its distal branches (LAD and circumflex), and coronary artery bypass grafting (CABG) using the left internal thoracic artery to the left anterior descending artery and a saphenous vein graft to a marginal branch. The patient’s postoperative course was complicated by transient multi-organ dysfunction, which resolved. He was discharged in a stable condition. This case highlights the critical importance of rapid multimodal imaging for diagnosis and the feasibility of emergency surgical intervention to achieve a favorable outcome in patients with a ruptured giant LMCA aneurysm. Full article

(This article belongs to the Collection Interesting Images)

► Show Figures

Figure 1

22 pages, 5732 KB

Open AccessArticle

Explainable Transformer-Based Framework for Glaucoma Detection from Fundus Images Using Multi-Backbone Segmentation and vCDR-Based Classification

by Hind Alasmari, Ghada Amoudi and Hanan Alghamdi

Diagnostics 2025, 15(18), 2301; https://doi.org/10.3390/diagnostics15182301 - 10 Sep 2025

Abstract

Glaucoma is an eye disease caused by increased intraocular pressure (IOP) that affects the optic nerve head (ONH), leading to vision problems and irreversible blindness. Background/Objectives: Glaucoma is the second leading cause of blindness worldwide, and the number of people affected is [...] Read more.

Glaucoma is an eye disease caused by increased intraocular pressure (IOP) that affects the optic nerve head (ONH), leading to vision problems and irreversible blindness. Background/Objectives: Glaucoma is the second leading cause of blindness worldwide, and the number of people affected is increasing each year, with the number expected to reach 111.8 million by 2040. This escalating trend is alarming due to the lack of ophthalmology specialists relative to the population. This study proposes an explainable end-to-end pipeline for automated glaucoma diagnosis from fundus images. It also evaluates the performance of Vision Transformers (ViTs) relative to traditional CNN-based models. Methods: The proposed system uses three datasets: REFUGE, ORIGA, and G1020. It begins with YOLOv11 for object detection of the optic disc. Then, the optic disc (OD) and optic cup (OC) are segmented using U-Net with ResNet50, VGG16, and MobileNetV2 backbones, as well as MaskFormer with a Swin-Base backbone. Glaucoma is classified based on the vertical cup-to-disc ratio (vCDR). Results: MaskFormer outperforms all models in segmentation in all aspects, including IoU OD, IoU OC, DSC OD, and DSC OC, with scores of 88.29%, 91.09%, 93.83%, and 93.71%. For classification, it achieved accuracy and F1-scores of 84.03% and 84.56%. Conclusions: By relying on the interpretable features of the vCDR, the proposed framework enhances transparency and aligns well with the principles of explainable AI, thus offering a trustworthy solution for glaucoma screening. Our findings show that Vision Transformers offer a promising approach for achieving high segmentation performance with explainable, biomarker-driven diagnosis. Full article

(This article belongs to the Section Machine Learning and Artificial Intelligence in Diagnostics)

► Show Figures

Figure 1

22 pages, 518 KB

Open AccessSystematic Review

Governing Artificial Intelligence in Radiology: A Systematic Review of Ethical, Legal, and Regulatory Frameworks

by Faten M. Aldhafeeri

Diagnostics 2025, 15(18), 2300; https://doi.org/10.3390/diagnostics15182300 - 10 Sep 2025

Abstract

Purpose: This systematic review explores the ethical, legal, and regulatory frameworks governing the deployment of artificial intelligence technologies in radiology. It aims to identify key governance challenges and highlight strategies that promote the safe, transparent, and accountable integration of artificial intelligence in clinical [...] Read more.

Purpose: This systematic review explores the ethical, legal, and regulatory frameworks governing the deployment of artificial intelligence technologies in radiology. It aims to identify key governance challenges and highlight strategies that promote the safe, transparent, and accountable integration of artificial intelligence in clinical imaging. This review is intended for both medical practitioners and AI developers, offering clinicians a synthesis of ethical and legal considerations while providing developers with regulatory insights and guidance for future AI system design. Methods: A systematic review was conducted, examining thirty-eight peer-reviewed articles published between 2018 and 2025. Studies were identified through searches in PubMed, Scopus, and Embase using terms related to artificial intelligence, radiology, ethics, law, and regulation. The inclusion criteria focused on studies addressing governance implications, rather than technical design. A thematic content analysis was applied to identify common patterns and gaps across ethical, legal, and regulatory domains. Results: The findings reveal widespread radiology-specific concerns, including algorithmic bias in breast and chest imaging datasets, opacity in image-based AI systems such as pulmonary nodule detection models, and unresolved legal liability in cases where radiologists rely on FDA-cleared AI tools that fail to identify abnormalities. Regulatory frameworks vary significantly across regions with limited global harmonization, highlighting the need for adaptive oversight models and improved data governance. Conclusion: Responsible deployment of AI in radiology requires governance models that address bias, explainability, and medico-legal accountability while integrating ethical principles, legal safeguards, and adaptive oversight. This review provides tailored insights for medical practitioners, AI developers, policymakers, and researchers: clinicians gain guidance on ethical and legal responsibilities, developers on regulatory and design priorities, policymakers (especially in the Middle East) on regional framework gaps, and researchers on future directions. Full article

(This article belongs to the Section Machine Learning and Artificial Intelligence in Diagnostics)

► Show Figures

Figure 1

26 pages, 11216 KB

Open AccessCase Report

Clinicopathological Pearls and Diagnostic Pitfalls in IgG4-Related Disease: Challenging Case Series and Literature Review

by Sokol Sina, Giulio Luigi Bonisoli, Sofia Vitale, Luigi Marzano, Stefano Francesco Crinò, Maria Cristina Conti Bellocchi, Sara Boninsegna, Simone Conci, Federica Maiolini, Riccardo Nocini, Luca Sacchetto, Giorgio Barbera, Andrea Fior, Nikela Kalaja, Elena Malloggi, Antonietta Brighenti, Alice Parisi, Nicolò Cardobi, Aldo Scarpa, Simonetta Friso and Elisa Tinazzi Show full author list Hide full author list

Diagnostics 2025, 15(18), 2299; https://doi.org/10.3390/diagnostics15182299 - 10 Sep 2025

Abstract

Background: IgG4-related disease (IgG4-RD) is a chronic immune-mediated fibroinflammatory disorder characterized by lymphoplasmacytic infiltrates enriched in IgG4-positive plasma cells, storiform fibrosis, and frequently elevated serum IgG4 levels. Classic forms, such as pancreaticobiliary or retroperitoneal involvement, are often recognized early, whereas atypical manifestations mimic [...] Read more.

Background: IgG4-related disease (IgG4-RD) is a chronic immune-mediated fibroinflammatory disorder characterized by lymphoplasmacytic infiltrates enriched in IgG4-positive plasma cells, storiform fibrosis, and frequently elevated serum IgG4 levels. Classic forms, such as pancreaticobiliary or retroperitoneal involvement, are often recognized early, whereas atypical manifestations mimic malignancy or inflammatory conditions, leading to delayed or inappropriate treatment. Case Series: A 30-year-old man presented with hyperemesis, proptosis, and gait instability. He was found to have colonic stenosis, stomach thickening, pachymeningitis, and polyserositis. Gastroenteric histology and serology confirmed IgG4-RD. Steroids were ineffective, but rituximab produced sustained clinical and radiologic improvement. A 35-year-old woman developed jaundice and cholestasis with a perihilar mass highly suggestive of cholangiocarcinoma. Histopathology revealed IgG4-RD, and rituximab therapy led to marked clinical and serological improvement. A 64-year-old woman with a submandibular mass underwent sialoadenectomy, with histology confirming IgG4-RD; she remained asymptomatic without systemic treatment. Literature Review: A literature review highlighted the diagnostic challenges of atypical IgG4-RD. Gastrointestinal involvement is rare and often misclassified as inflammatory bowel disease. Isolated biliary disease frequently mimics cholangiocarcinoma, while salivary gland involvement may be misdiagnosed as neoplasia. Serum IgG4 levels >135 mg/dL and IgG4/IgG ratio >0.21 may support clinical suspicion, but histopathology remains indispensable for definitive diagnosis and for excluding malignancy. Steroid responsiveness is a hallmark, though relapses after tapering are common, often necessitating B-cell-directed therapy. Conclusions: IgG4-RD should be considered in patients with unexplained, relapsing, or steroid-responsive conditions. Early recognition, multidisciplinary collaboration, and integration of histopathology with clinical features are essential to avoid misdiagnosis and optimize management. Full article

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

► Show Figures

Figure 1

18 pages, 1552 KB

Open AccessArticle

Comparative Evaluation of 24 LDL-C Estimation Equations Against Direct Assays in Two Independent Cohorts

by Imola Györfi, Oana Roxana Oprea, Ion Bogdan Mănescu, Antoanela Curici and Minodora Dobreanu

Diagnostics 2025, 15(18), 2298; https://doi.org/10.3390/diagnostics15182298 - 10 Sep 2025

Abstract

Background: Low-density lipoprotein cholesterol (LDL-C) is essential in diagnosing and managing dyslipidemias. While direct assays are faster than the reference beta-quantification method, many labs continue using the Friedewald (FW) equation, despite its limitations. Methods: Two large datasets were analyzed: 10,174 hospital samples (Cobas/Roche) [...] Read more.

Background: Low-density lipoprotein cholesterol (LDL-C) is essential in diagnosing and managing dyslipidemias. While direct assays are faster than the reference beta-quantification method, many labs continue using the Friedewald (FW) equation, despite its limitations. Methods: Two large datasets were analyzed: 10,174 hospital samples (Cobas/Roche) and 21,091 private lab samples (Alinity/Abbott). Various literature-based LDL-C equations were compared, focusing on FW, Sampson (SN), and Martin–Hopkins (MH). Direct LDL-C served as the reference. Evaluation metrics included bias and classification accuracy. Results: In samples with triglycerides < 400 mg/dL, several lesser-known equations showed acceptable bias (±5%), outperforming FW, SN, and MH, which had biases from −7.4% to −4.9%. Classification accuracy was higher with equations like Vujovic (up to 82.5%), compared to FW (65.8%), SN (73.1%), and MH (72.4%). The Vujovic equation showed minimal bias across triglyceride levels and the highest net gain in correct classification (3.4% and 1.57%). Conclusions: Multiple lesser-known LDL-C formulas outperformed widely used ones. The Vujovic equation yielded the best results, but the limited net clinical improvement suggests that replacing Friedewald may not be urgently necessary. Full article

(This article belongs to the Special Issue Exploring the Role of Diagnostic Biochemistry, 2nd Edition)

► 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