DiagnosticsPub Date : 2025-03-12DOI: 10.3390/diagnostics15060709
Hiroko Naganuma, Hideaki Ishida
{"title":"Demonstration of Hepatic Vein Abnormalities Using Contrast-Enhanced Sonography in Liver Diseases.","authors":"Hiroko Naganuma, Hideaki Ishida","doi":"10.3390/diagnostics15060709","DOIUrl":"10.3390/diagnostics15060709","url":null,"abstract":"<p><p>Contrast-enhanced US (CEUS) is now widely used to observe the hemodynamics of the liver. The CEUS diagnosis mainly consists of evaluating hepatic artery and portal vein flow changes in liver diseases, but it has not been widely used for the diagnosis of hepatic venous (HV) abnormalities in the clinical setting. For this background, this review tried to reconsider this problem. In short, observing HV CEUS findings, especially HV transit time, serves to largely narrow the differential diagnosis and increase the diagnostic confidence of the CEUS. However, diagnosing HV CEUS diagnosis in a wide range of liver diseases requires understanding of vascular anatomy of the upper abdomen and vascular structure of each disease. Additionally, interpreting CEUS findings of HCC should be prudent, because its drainage vessels change according to the histological progression, from the HV to the portal vein. Thus, the most important way of making use of the CEUS information is interpreting it in conjunction with the clinical data.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 6","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11941399/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143729129","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-03-12DOI: 10.3390/diagnostics15060703
Pei-An Chen, Chiu-Hsuan Cheng, Dah-Ching Ding
{"title":"The Coincidence of Ovarian Endometrioma with Paratubal Leydig Cell Nodules: A Case Report and Literature Review.","authors":"Pei-An Chen, Chiu-Hsuan Cheng, Dah-Ching Ding","doi":"10.3390/diagnostics15060703","DOIUrl":"10.3390/diagnostics15060703","url":null,"abstract":"<p><p><b>Background and Clinical Significance:</b> Paratubal Leydig cell nodules are rare incidental findings that present diagnostic challenges. <b>Case Presentation</b>: A 45-year-old female with a history of hypertension and diabetes mellitus presented with fever and chills following an episode of severe dysmenorrhea and menorrhagia. The patient reported heavy menstrual bleeding, persisting for 2-3 years. Physical examination revealed erythema of the perineum and whitish vaginal discharge, with no cervical lesions. Imaging revealed a 15 cm right ovarian cyst. Laboratory investigations showed elevated C-reactive protein (6.37 mg/L) and CA125 (88.82 U/mL) levels, whereas other tumor markers were within normal limits. A pelvic ultrasound revealed a retroverted uterus and a large ovarian mass suggestive of malignancy. The patient underwent a right salpingo-oophorectomy, during which a 15 cm ovarian tumor adherent to the right pelvic sidewall was excised. Histopathological examination revealed an endometriotic cyst with endometrial glandular epithelium positive for estrogen receptor and focal mucinous metaplasia. CD10-positive endometrial stromal cells and paratubal cysts were also observed. Additionally, a small Leydig cell tumor originated from the ovarian hilum was identified and confirmed by positive staining for inhibin, calretinin, and androgen receptors, as well as negative estrogen receptor staining. The postoperative recovery was uneventful, and at the five-week follow-up, the patient's hormonal levels were normal, and there were no complications. <b>Conclusions</b>: This case highlights the importance of thorough histopathological evaluation in managing ovarian masses and the potential coexistence of benign and rare pathological entities, such as Leydig cell tumors.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 6","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11941348/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143729254","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-03-12DOI: 10.3390/diagnostics15060699
Farid Ziayee, Hannan Dalyanoglu, Christian Schnitzler, Kai Jannusch, Matthias Boschheidgen, Judith Boeven, Hug Aubin, Bernd Turowski, Marius Georg Kaschner, Christian Mathys
{"title":"A Retrospective Analysis of the Effects of Concomitant Use of Intra-Aortic Balloon Pump (IABP) and Veno-Arterial Extracorporeal Membrane Oxygenation (va-ECMO) Therapy on Procedural Brain Infarction.","authors":"Farid Ziayee, Hannan Dalyanoglu, Christian Schnitzler, Kai Jannusch, Matthias Boschheidgen, Judith Boeven, Hug Aubin, Bernd Turowski, Marius Georg Kaschner, Christian Mathys","doi":"10.3390/diagnostics15060699","DOIUrl":"10.3390/diagnostics15060699","url":null,"abstract":"<p><p><b>Background/Objectives:</b> Brain ischemia is a frequent complication in patients undergoing veno-arterial extracorporeal membrane oxygenation (va-ECMO) therapy due to hypoperfusion, low oxygenation, and thromboembolism. While concomitant intra-aortic balloon pump (IABP) therapy may improve the perfusion of the supra-aortic branches, it may also favor thromboembolism. This retrospective study aimed to evaluate the effects of combined va-ECMO and IABP therapy on procedural brain infarction compared to va-ECMO therapy alone, with a specific focus on analyzing the types of infarctions. <b>Methods:</b> Cranial computed tomography (CCT) scans of consecutive patients receiving va-ECMO therapy were analyzed retrospectively. Subgroups were formed for patients with combined therapy (ECMO and IABP) and va-ECMO therapy only. The types of infarctions and the potential impacts of va-ECMO vs. combined therapy with IABP on stroke were investigated. <b>Results:</b> Overall, 146 patients (36 female, 110 male, mean age 61 ± 13.3 years) were included, with 69 undergoing combined therapy and 77 patients receiving va-ECMO therapy alone. In total, 14 stroke events occurred in 11 patients in the ECMO-only group and there were 12 events in 12 patients in the ECMO + IABP-group, showing no significant difference (<i>p</i> = 0.61). The majority of infarctions were of thromboembolic (<i>n</i> = 23; 88%) origin, with 14 stroke-events in 12 patients in the ECMO + IABP-group and 9 stroke events in the ECMO-only group. The survival rate within 30 days of treatment was 29% in the ECMO-only group and 32% in the ECMO + IABP group. <b>Conclusions:</b> The results of this retrospective study show that concomitant IABP therapy appears to be neither protective nor more hazardous in relation to ECMO-related stroke. Thus, the indication for additional IABP therapy should be assessed independently from the procedural risk of brain ischemia. Thromboembolic infarctions seem to represent the most common type of infarction in ECMO, especially within the first 48 h of treatment.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 6","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11940886/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143729118","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-03-12DOI: 10.3390/diagnostics15060710
Olusegun Ekundayo Adebayo, Brice Chatelain, Dumitru Trucu, Raluca Eftimie
{"title":"Deep Learning Approaches for the Classification of Keloid Images in the Context of Malignant and Benign Skin Disorders.","authors":"Olusegun Ekundayo Adebayo, Brice Chatelain, Dumitru Trucu, Raluca Eftimie","doi":"10.3390/diagnostics15060710","DOIUrl":"10.3390/diagnostics15060710","url":null,"abstract":"<p><p><b>Background/Objectives:</b> Misdiagnosing skin disorders leads to the administration of wrong treatments, sometimes with life-impacting consequences. Deep learning algorithms are becoming more and more used for diagnosis. While many skin cancer/lesion image classification studies focus on datasets containing dermatoscopic images and do not include keloid images, in this study, we focus on diagnosing keloid disorders amongst other skin lesions and combine two publicly available datasets containing non-dermatoscopic images: one dataset with keloid images and one with images of other various benign and malignant skin lesions (melanoma, basal cell carcinoma, squamous cell carcinoma, actinic keratosis, seborrheic keratosis, and nevus). <b>Methods:</b> Different Convolution Neural Network (CNN) models are used to classify these disorders as either malignant or benign, to differentiate keloids amongst different benign skin disorders, and furthermore to differentiate keloids among other similar-looking malignant lesions. To this end, we use the transfer learning technique applied to nine different base models: the VGG16, MobileNet, InceptionV3, DenseNet121, EfficientNetB0, Xception, InceptionRNV2, EfficientNetV2L, and NASNetLarge. We explore and compare the results of these models using performance metrics such as accuracy, precision, recall, F1<sub><i>score</i></sub>, and AUC-ROC. <b>Results:</b> We show that the VGG16 model (after fine-tuning) performs the best in classifying keloid images among other benign and malignant skin lesion images, with the following keloid class performance: an accuracy of 0.985, precision of 1.0, recall of 0.857, F1 score of 0.922 and AUC-ROC value of 0.996. VGG16 also has the best overall average performance (over all classes) in terms of the AUC-ROC and the other performance metrics. Using this model, we further attempt to predict the identification of three new non-dermatoscopic anonymised clinical images, classifying them as either malignant, benign, or keloid, and in the process, we identify some issues related to the collection and processing of such images. Finally, we also show that the DenseNet121 model has the best performance when differentiating keloids from other malignant disorders that have similar clinical presentations. <b>Conclusions:</b> The study emphasised the potential use of deep learning algorithms (and their drawbacks), to identify and classify benign skin disorders such as keloids, which are not usually investigated via these approaches (as opposed to cancers), mainly due to lack of available data.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 6","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11940829/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143729152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-03-12DOI: 10.3390/diagnostics15060708
Hritvik Jain, Maryam Shahzad, Muneeba Ahsan, Rahul Patel, Jagjot Singh, Ramez M Odat, Aman Goyal, Raveena Kelkar, Nishad Barve, Hina Farrukh, Raheel Ahmed
{"title":"Diagnostic Value of Comprehensive Echocardiographic Assessment Including Speckle-Tracking in Patients with Sarcoidosis Versus Healthy Controls: A Systematic Review and Meta-Analysis.","authors":"Hritvik Jain, Maryam Shahzad, Muneeba Ahsan, Rahul Patel, Jagjot Singh, Ramez M Odat, Aman Goyal, Raveena Kelkar, Nishad Barve, Hina Farrukh, Raheel Ahmed","doi":"10.3390/diagnostics15060708","DOIUrl":"10.3390/diagnostics15060708","url":null,"abstract":"<p><p><b>Background</b>: Cardiac involvement in sarcoidosis is often subclinical, with late manifestations associated with poorer prognosis. Speckle-tracking echocardiography (STE) is gaining attention due to its ability to detect subclinical alterations in myocardial contraction patterns and quantification of abnormal parameters. <b>Methods</b>: Databases, including PubMed, Cochrane Central, Embase, Scopus, and Web of Science, were searched to identify studies comparing echocardiographic parameters in sarcoidosis patients with healthy controls. Mean difference (MD) with 95% confidence intervals (CI) were pooled using the inverse-variance random-effects model in Review Manager Version 5.4.1. Statistical significance was considered at <i>p</i>-value <0.05. <b>Results</b>: Thirteen studies with 1416 participants (854-sarcoidosis; 562-healthy controls) were included. In a pooled analysis, patients with sarcoidosis demonstrated a significantly lower left ventricular global longitudinal strain (LV GLS) (Mean Difference [MD]: -3.60; 95% Confidence Interval [CI]: -4.76, -2.43; <i>p</i> < 0.0001) and left ventricular global circumferential strain (LV GCS) (MD: -2.52; 95% CI: -4.61, -0.43; <i>p</i> = 0.02), along with a significantly higher pulmonary artery systolic pressure (PASP) (MD: 4.19; 95% CI: 0.08, 8.29; <i>p</i> = 0.05), left ventricular end-systolic diameter (LVESD) (MD: 0.90; 95% CI: 0.10, 1.71; <i>p</i> = 0.03), A-wave velocity (MD: 3.36; 95% CI: 0.33, 6.39; <i>p</i> = 0.03), and E/E' ratio (MD: 1.33; 95% CI: 0.42, 2.23; <i>p</i> = 0.004) compared to healthy controls. No significant differences were noted in left ventricular ejection fraction (LVEF), left ventricular global radial strain (LV GRS), interventricular septal thickness (IVST), tricuspid annular plane systolic excursion (TAPSE), left ventricular end-diastolic diameter (LVEDD), E-wave velocity, and E/A ratio. <b>Conclusions</b>: STE serves as a promising imaging modality in detecting subclinical cardiac involvement in sarcoidosis patients with no overt cardiac manifestations. A widespread cardiovascular evaluation of sarcoidosis patients with STE is recommended to detect these altered myocardial contractile patterns. The early detection of cardiac sarcoidosis is essential to prevent adverse clinical outcomes and improve mortality.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 6","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11941600/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143729285","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Contrast Volume Reduction in Oncologic Body Imaging Using Dual-Energy CT: A Comparison with Single-Energy CT.","authors":"Marianna Gulizia, Anais Viry, Mario Jreige, Guillaume Fahrni, Yannick Marro, Gibran Manasseh, Christine Chevallier, Clarisse Dromain, Naik Vietti-Violi","doi":"10.3390/diagnostics15060707","DOIUrl":"10.3390/diagnostics15060707","url":null,"abstract":"<p><p><b>Background/Objectives</b>: To evaluate the feasibility of reducing contrast volume in oncologic body imaging using dual-energy CT (DECT) by (1) identifying the optimal virtual monochromatic imaging (VMI) reconstruction using DECT and (2) comparing DECT performed with reduced iodinated contrast media (ICM) volume to single-energy CT (SECT) performed with standard ICM volume. <b>Methods</b>: In this retrospective study, we quantitatively and qualitatively compared the image quality of 35 thoracoabdominopelvic DECT across 9 different virtual monoenergetic image (VMI) levels (from 40 to 80 keV) using a reduced volume of ICM (0.3 gI/kg of body weight) to determine the optimal keV reconstruction level. Out of these 35 patients, 20 had previously performed SECT with standard ICM volume (0.3 gI/kg of body weight + 9 gI), enabling protocol comparison. The qualitative analysis included overall image quality, noise, and contrast enhancement by two radiologists. Quantitative analysis included contrast enhancement measurements, contrast-to-noise ratio, and signal-to-noise ratio of the liver parenchyma and the portal vein. ANOVA was used to identify the optimal VMI level reconstruction, while <i>t</i>-tests and paired <i>t</i>-tests were used to compare both protocols. <b>Results</b>: VMI<sub>60 keV</sub> provided the highest overall image quality score. DECT with reduced ICM volume demonstrated higher contrast enhancement and lower noise than SECT with standard ICM volume (<i>p</i> < 0.001). No statistical difference was found in the overall image quality between the two protocols (<i>p</i> = 0.290). <b>Conclusions</b>: VMI<sub>60 keV</sub> with reduced contrast volume provides higher contrast and lower noise than SECT at a standard contrast volume. DECT using a reduced ICM volume is the technique of choice for oncologic body CT.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 6","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11941575/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143729085","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-03-11DOI: 10.3390/diagnostics15060686
Juan Jesús Fernández Alba, Florentino Carral, Carmen Ayala Ortega, Jose Diego Santotoribio, María Castillo Lara, Carmen González Macías
{"title":"External Validation of a Predictive Model for Thyroid Cancer Risk with Decision Curve Analysis.","authors":"Juan Jesús Fernández Alba, Florentino Carral, Carmen Ayala Ortega, Jose Diego Santotoribio, María Castillo Lara, Carmen González Macías","doi":"10.3390/diagnostics15060686","DOIUrl":"10.3390/diagnostics15060686","url":null,"abstract":"<p><p><b>Background/Objectives</b>: Thyroid cancer ranks among the most prevalent endocrine neoplasms, with a significant rise in incidence observed in recent decades, particularly in papillary thyroid carcinoma (PTC). This increase is largely attributed to the enhanced detection of subclinical cancers through advanced imaging techniques and fine-needle aspiration biopsies. The present study aims to externally validate a predictive model previously developed by our group, designed to assess the risk of a thyroid nodule being malignant. <b>Methods</b>: By utilizing clinical, analytical, ultrasound, and histological data from patients treated at the Puerto Real University Hospital, this study seeks to evaluate the performance of the predictive model in a distinct dataset and perform a decision curve analysis to ascertain its clinical utility. <b>Results</b>: A total of 455 patients with thyroid nodular pathology were studied. Benign nodular pathology was diagnosed in 357 patients (78.46%), while 98 patients (21.54%) presented with a malignant tumor. The most frequent histological type of malignant tumor was papillary cancer (71.4%), followed by follicular cancer (6.1%). Malignant nodules were predominantly solid (95.9%), hypoechogenic (72.4%), with irregular or microlobed borders (36.7%), and associated with suspicious lymph nodes (24.5%). The decision curve analysis confirmed the model's accuracy and its potential impact on clinical decision-making. <b>Conclusions</b>: The external validation of our predictive model demonstrates its robustness and generalizability across different populations and clinical settings. The integration of advanced diagnostic tools, such as AI and ML models, improves the accuracy in distinguishing between benign and malignant nodules, thereby optimizing treatment strategies and minimizing invasive procedures. This approach not only facilitates the early detection of cancer but also helps to avoid unnecessary surgeries and biopsies, ultimately reducing patient morbidity and healthcare costs.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 6","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11941551/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143728857","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Flash Glucose Monitoring for Predicting Cardiogenic Shock Occurrence in Critically Ill Patients: A Retrospective Pilot Study.","authors":"Velimir Altabas, Dorijan Babić, Anja Grulović, Tomislav Bulum, Zdravko Babić","doi":"10.3390/diagnostics15060685","DOIUrl":"10.3390/diagnostics15060685","url":null,"abstract":"<p><p><b>Background/Objectives:</b> Continuous and flash glucose monitoring (CGM and FGM) may enhance glucose management by providing real-time glucose data. Furthermore, growing evidence is linking altered blood glucose concentrations and worse short-term outcomes in critically ill patients. While hyperglycemia is more common in these patients and is associated with an increased risk of adverse events, hypoglycemia is particularly concerning and significantly raises the risk of fatal outcomes. This exploratory study investigated the link between FGM variables and cardiogenic shock in critically ill Coronary Care Unit (CCU) patients. <b>Methods:</b> Twenty-eight CCU patients (1 May 2021-31 January 2022) were monitored using a Libre FreeStyle system. Analyzed data included patient demographic and laboratory data, left ventricular ejection fraction, standard glucose monitoring, APACHE IV scores, and cardiogenic shock occurrence. Analysis was performed using the χ<sup>2</sup> test, Mann-Whitney U test, and logistic regression. <b>Results:</b> Among the patients, 13 (46.43%) developed cardiogenic shock. FGM detected hypoglycemia in 18 (64.29%) patients, while standard methods in 6 (21.43%) patients. FGM-detected hypoglycemia was more frequent in patients who developed cardiogenic shock (<i>p</i> = 0.0129, χ<sup>2</sup> test) with a significantly higher time below range reading (<i>p</i> = 0.0093, Mann Withney U test), despite no differences in mean glucose values. In addition, hypoglycemia detected by FGM was an independent predictor of shock (<i>p</i> = 0.0390, logistic regression). <b>Conclusions:</b> FGM identified more hypoglycemic events compared to standard glucose monitoring in the CCU. Frequent FGM-detected hypoglycemic events were associated with cardiogenic shock, regardless of a history of diabetes. Due to a limited sample size, these results should be interpreted cautiously and further research in this area is justified.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 6","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11941065/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143728868","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-03-11DOI: 10.3390/diagnostics15060689
Aanuoluwapo Clement David-Olawade, David B Olawade, Laura Vanderbloemen, Oluwayomi B Rotifa, Sandra Chinaza Fidelis, Eghosasere Egbon, Akwaowo Owoidighe Akpan, Sola Adeleke, Aruni Ghose, Stergios Boussios
{"title":"AI-Driven Advances in Low-Dose Imaging and Enhancement-A Review.","authors":"Aanuoluwapo Clement David-Olawade, David B Olawade, Laura Vanderbloemen, Oluwayomi B Rotifa, Sandra Chinaza Fidelis, Eghosasere Egbon, Akwaowo Owoidighe Akpan, Sola Adeleke, Aruni Ghose, Stergios Boussios","doi":"10.3390/diagnostics15060689","DOIUrl":"10.3390/diagnostics15060689","url":null,"abstract":"<p><p>The widespread use of medical imaging techniques such as X-rays and computed tomography (CT) has raised significant concerns regarding ionizing radiation exposure, particularly among vulnerable populations requiring frequent imaging. Achieving a balance between high-quality diagnostic imaging and minimizing radiation exposure remains a fundamental challenge in radiology. Artificial intelligence (AI) has emerged as a transformative solution, enabling low-dose imaging protocols that enhance image quality while significantly reducing radiation doses. This review explores the role of AI-assisted low-dose imaging, particularly in CT, X-ray, and magnetic resonance imaging (MRI), highlighting advancements in deep learning models, convolutional neural networks (CNNs), and other AI-based approaches. These technologies have demonstrated substantial improvements in noise reduction, artifact removal, and real-time optimization of imaging parameters, thereby enhancing diagnostic accuracy while mitigating radiation risks. Additionally, AI has contributed to improved radiology workflow efficiency and cost reduction by minimizing the need for repeat scans. The review also discusses emerging directions in AI-driven medical imaging, including hybrid AI systems that integrate post-processing with real-time data acquisition, personalized imaging protocols tailored to patient characteristics, and the expansion of AI applications to fluoroscopy and positron emission tomography (PET). However, challenges such as model generalizability, regulatory constraints, ethical considerations, and computational requirements must be addressed to facilitate broader clinical adoption. AI-driven low-dose imaging has the potential to revolutionize radiology by enhancing patient safety, optimizing imaging quality, and improving healthcare efficiency, paving the way for a more advanced and sustainable future in medical imaging.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 6","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11941271/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143729132","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DiagnosticsPub Date : 2025-03-11DOI: 10.3390/diagnostics15060691
Saleh Albahli
{"title":"A Robust YOLOv8-Based Framework for Real-Time Melanoma Detection and Segmentation with Multi-Dataset Training.","authors":"Saleh Albahli","doi":"10.3390/diagnostics15060691","DOIUrl":"10.3390/diagnostics15060691","url":null,"abstract":"<p><p><b>Background</b>: Melanoma, the deadliest form of skin cancer, demands accurate and timely diagnosis to improve patient survival rates. However, traditional diagnostic approaches rely heavily on subjective clinical interpretations, leading to inconsistencies and diagnostic errors. <b>Methods</b>: This study proposes a robust YOLOv8-based deep learning framework for real-time melanoma detection and segmentation. A multi-dataset training strategy integrating the ISIC 2020, HAM10000, and PH2 datasets was employed to enhance generalizability across diverse clinical conditions. Preprocessing techniques, including adaptive contrast enhancement and artifact removal, were utilized, while advanced augmentation strategies such as CutMix and Mosaic were applied to enhance lesion diversity. The YOLOv8 architecture unified lesion detection and segmentation tasks into a single inference pass, significantly enhancing computational efficiency. <b>Results</b>: Experimental evaluation demonstrated state-of-the-art performance, achieving a mean Average Precision (mAP@0.5) of 98.6%, a Dice Coefficient of 0.92, and an Intersection over Union (IoU) score of 0.88. These results surpass conventional segmentation models including U-Net, DeepLabV3+, Mask R-CNN, SwinUNet, and Segment Anything Model (SAM). Moreover, the proposed framework demonstrated real-time inference speeds of 12.5 ms per image, making it highly suitable for clinical deployment and mobile health applications. <b>Conclusions</b>: The YOLOv8-based framework effectively addresses the limitations of existing diagnostic methods by integrating detection and segmentation tasks, achieving high accuracy and computational efficiency. This study highlights the importance of multi-dataset training for robust generalization and recommends the integration of explainable AI techniques to enhance clinical trust and interpretability.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 6","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11941149/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143729140","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}