Non-invasive Tool Research Articles

GEARBOCS: An Adeno Associated Virus Tool for In Vivo Gene Editing in Astrocytes.

CRISPR/Cas9-based genome engineering enables rapid and precise gene manipulations in the CNS. Here, we developed a non-invasive astrocyte-specific method utilizing a single AAV vector, which we named GEARBOCS (Gene Editing in AstRocytes Based On CRISPR/Cas9 System). We verified GEARBOCS' specificity to mouse cortical astrocytes and demonstrated its utility for three types of gene manipulations: knockout (KO); tagging (TagIn); and reporter knock-in (GeneTrap) strategies. Next, we deployed GEARBOCS in two test cases. First, we determined that astrocytes are a necessary source of the synaptogenic factor Sparcl1 for thalamocortical synapse maintenance in the mouse primary visual cortex. Second, we determined that cortical astrocytes express the synaptic vesicle associated Vamp2 protein and found that it is required for maintaining excitatory and inhibitory synapse numbers in the visual cortex. These results show that the GEARBOCS strategy provides a fast and efficient means to study astrocyte biology in vivo.

Integrating proximal geophysical sensing and machine learning for digital soil mapping: Spatial prediction and model evaluation using a small dataset

Geophysical methods support soil security by providing non-invasive tools to assess soil properties, monitor degradation, and guide sustainable management strategies. However, studies focusing the spatial prediction of geophysical data remain limited. In this research, we aimed to model and predict the spatial distribution of soil geophysical properties using parent material and terrain attributes with machine learning algorithms. In addition, we tested the nested leave-one-out cross validation (nested-LOOCV) method to deal with datasets with limited size. We performed a geophysical survey using three types of sensors (radiometric, magnetic and electric methods). The random forest (RF) and support vector machine (SVM) algorithms presented the best results, with RF showing higher performance for K40 and magnetic susceptibility, and SVM had higher performance for eU, eTh and apparent electrical conductivity. Parent materials and digital elevation model were the most significant variables for the modelling. The nested-LOOCV method proved to be adequate for small soil dataset. Machine learning techniques are potential tools for modelling soil geophysical variables. The combination with computational techniques shows the great relevance of geophysical measurements for the estimation of soil properties related to fertility and soil genesis.

Severity Stratification of Coronary Artery Disease Using Novel Inner Ellipse-Based Foveal Avascular Zone Biomarkers.

Given the similarities between the retinal and coronary microvasculature, the retina holds promising potential to serve as a non-invasive screening tool for coronary artery disease (CAD). We aimed to develop novel inner ellipse-based metrics and discern whether foveal avascular zone (FAZ) alterations can serve as indicators for CAD presence and severity. Patients admitted to the Department of Cardiology who underwent coronary angiography were included. This resulted in an inclusion of 212 patients, of which 73 had no CAD. During the same visit, 6 × 6-mm (nominal size) fovea-centered optical coherence tomography angiography images of both eyes were acquired. The Gensini score (GS) was utilized to quantify CAD severity. Six known FAZ shape metrics were assessed and three novel biomarkers based on the inner ellipse were defined: absolute inner ellipse difference, Hausdorff distance, and Chamfer distance. Eight out of nine metrics showed significant associations with the GS in the left eye. However, significant differences across three CAD severity groups were only demonstrated by the novel metrics. Utilizing the Chamfer distance, age, and sex, patients with and without CAD could be distinguished with an average area under the curve (AUC) of 0.89 (95% confidence interval [CI], 0.84-0.95). Moreover, three CAD severity groups could be discerned with a macro average AUC of 0.77 (95% CI, 0.72-0.84). A comprehensive assessment of FAZ shape descriptors was performed, and a strong association with CAD was found. The inner ellipse-based biomarkers especially demonstrated high predictive abilities for CAD presence and severity.

Metabolic Rate of Goldfish (Carassius auratus) in the Face of Common Aquaculture Challenges.

This study examined the metabolic rate (MO2, oxygen consumption) of goldfish (Carassius auratus) under normal management conditions in aquaculture. Using an intermittent respirometry system, we assessed daily variations and the effects of feeding, handling, temperature increase, and anesthetics. MO2 exhibited a daily rhythm, with higher values during day. Feeding to satiety produced a 35% increase in MO2 compared to fasted animals, with a maximum peak after 3 h and returning to baseline after 7 h. Handling stress (5 min) produced a 140% MO2 peak (from 180 to 252 mg O2 kg-1 h-1), returning to the routine MO2 after 2.5 h. An increase in water temperature (+0.1 °C min-1) up to 30 °C caused MO2 to peak at 200% after 2.5 h from the start of the temperature increase. The use of common anesthetics in aquaculture (MS-222, 2-phenoxyethanol and clove oil in deep anesthesia concentration) affects MO2 during the first few minutes after anesthetic recovery, but also during the following 4 h. It can be concluded that the metabolic rate is a good indicator of the goldfish's response to aquaculture practices involving energy expenditure and stress. Thus, intermittent respirometry is a valuable non-invasive tool for understanding and improving fish welfare in aquaculture.

Predictive Accuracy of C-reactive Protein as an Early Indicator of Leakage After Anastomosis and Primary Repair in Gastrointestinal Surgery.

Background Anastomotic leakage after gastrointestinal surgery is a significant postoperative complication that leads to increased morbidity and mortality. C-reactive protein (CRP) has been used previously as a predictive marker of anastomotic leakage in gastrointestinal surgeries. Its short half-life makes it a reliable marker for postoperative complications, rapidly returning to normal values as the patient recovers. We conducted this study to evaluate the predictive accuracy of serum CRP levels in predicting anastomotic leaks in patients undergoing gastrointestinal repair surgeries. Methods Ninety-six gastrointestinal surgeries involving anastomoses and primary repairs were included in the study. CRP was taken serially from postoperative days (POD) 1 to 7. Patients were divided into two groups based on postoperative outcomes: leakage and non-leakage. The receiver operating characteristic (ROC) curve of CRP levels with leak and mortality was plotted to find a threshold value for leak and mortality. Results Out of 96 patients included in the study, Group B (non-leakage) consisted of 78 patients (81.3%), while Group A (leakage) comprised 18 patients (18.7%). ROC analysis identified a CRP cutoff of 127 mg/L on POD 5, with high sensitivity (80%) and high specificity (80%) indicating a high likelihood of leakage above this threshold. Conclusion This study underscores the importance of monitoring CRP levels in the postoperative period, particularly on POD 5, as a non-invasive and cost-effective tool for the early detection of anastomotic leaks.

Machine learning-based diagnostic prediction of minimal change disease: model development study

Minimal change disease (MCD) is a common cause of nephrotic syndrome. Due to its rapid progression, early detection is essential; however, definitive diagnosis requires invasive kidney biopsy. This study aims to develop non-invasive predictive models for diagnosing MCD by machine learning. We retrospectively collected data on demographic characteristics, blood tests, and urine tests from patients with nephrotic syndrome who underwent kidney biopsy. We applied four machine learning algorithms—TabPFN, LightGBM, Random Forest, and Artificial Neural Network—and logistic regression. We compared their performance using stratified 5-repeated 5-fold cross-validation for the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC). Variable importance was evaluated using the SHapley Additive exPlanations (SHAP) method. A total of 248 patients were included, with 82 cases (33%) were diagnosed with MCD. TabPFN demonstrated the best performance with an AUROC of 0.915 (95% CI 0.896–0.932) and an AUPRC of 0.840 (95% CI 0.807–0.872). The SHAP methods identified C3, total cholesterol, and urine red blood cells as key predictors for TabPFN, consistent with previous reports. Machine learning models could be valuable non-invasive diagnostic tools for MCD.

Chemoradiotherapy treatment increases cardiac and aortic [18F]FDG uptake ratios in lung cancer patients

[18F]Fluorodeoxyglucose (FDG) positron emission tomography (PET) is an indispensable non-invasive imaging tool to aid diagnosis, prognostication, and therapeutic monitoring in oncology, but it can also evaluate cardiovascular inflammation1,2. Previously, cardiac metabolic changes using [18F]FDG with chemoradiotherapy have been explored3 but changes in the rest of the cardiovascular system remain undetermined. To investigate the cardiovascular metabolic changes pre- and post-chemoradiotherapy in stage 3b non-small cell lung cancer (NSCLC) patients. A retrospective analysis of 26 patients (43-82 years) with stage 3b NSCLC from the American College of Radiology Imaging Network (ACRIN 6668) trial was performed3. All available pre- and post-treatment imaging were retrieved from the Cancer Imaging Archive (TCIA)4 and regions of interest for the left ventricle and calcified large arteries were contoured on all PET-CT scans using PMOD version 4.0 software. The mean, maximum standard uptake value (SUVmean and SUVmax) and target-to-background ratio (TBR) were quantified on PMOD. At approximately 14 weeks post-chemoradiotherapy, there was a higher SUVmean (mean difference 0.81, p=0.01) and SUVmax (mean difference 2.20, p=0.006) in the left ventricle compared with pretreatment scans. TBR for aorta was higher post-treatment (mean difference 0.06, p=0.005). However, SUVmean for carotid arteries and right brachiocephalic artery was reduced post-chemoradiotherapy (mean difference 0.15, p=0.008 and mean difference 0.28, p=0.03). A reduction in SUVmax was also seen for aortic arch (mean difference 0.58, p=0.03) and right brachiocephalic artery (mean difference 0.42, p=0.03 and mean difference 0.42, p=0.03). Cardiovascular glucose metabolism is selectively increased in the left ventricle and aorta post-chemoradiotherapy. Changes in glucose metabolism of atherosclerotic plaques vary across different vessels throughout the body. Please click on the 'PDF' for the full abstract!

An updated approach to the evaluation of the urinary sediment.

Examination of the urinary sediment (U-sed) is an important non-invasive, rapid, and inexpensive tool for the diagnosis and surveillance over time of renal diseases. In this Educational Review, we describe first how to collect, prepare, and examine urine samples in order to obtain reliable results. Then, we describe the U-sed findings in isolated microscopic hematuria, glomerular diseases, acute interstitial nephritis, acute kidney injury, reactivation of the BK virus in kidney transplant recipients, and crystalluric genetic diseases.

Evaluation of a plasma cell-free DNA methylation test for colorectal cancer diagnosis: a multicenter clinical study.

A blood-based diagnostic test is a promising strategy for colorectal cancer (CRC). The MethyDT test (IColohunter), which detects methylation levels of NTMT1 and MAP3K14-AS1, exhibited potential in discriminating CRC, but its clinical performance needs to be validated in large-scale populations. A multicenter, double-blinded, cross-sectional study that enrolled 1194 participants was performed. Plasma samples were collected to detect methylation levels of NTMT1 and MAP3K14-AS1 using quantitative methylation-specific PCR with the MethyDT test, and the accuracy was further evaluated by Sanger sequencing. The sensitivities of the MethyDT test for detecting CRC, early stages of CRC (I and II), advanced adenoma (AA), and high-grade intraepithelial neoplasia (HGIN) were 91.2% (95% confidence interval [CI], 88.4-94.0), 87.4% (95% CI, 82.5-92.2), 43.5% (95% CI, 35.7-51.4), and 72.7% (95% CI, 57.5-87.9), respectively. The specificities for participants with non-AA, interfering diseases (ID), and no evidence of disease (NED) were 85.0% (95% CI, 78.8-91.3), 93.7% (95% CI, 91.4-95.9) and 97.3% (95% CI, 90.5-99.7), respectively, and its overall specificity for all-controls was 92.4% (95% CI, 90.3-94.4). The consistency of the MethyDT test with pathology for CRC was high with a kappa value of 0.830 (95% CI, 0.795-0.865). Additionally, the MethyDT test was comparable to Sanger sequencing for detecting methylation with kappa values > 0.97. The MethyDT test demonstrates excellent sensitivity and specificity for CRC and high consistency with Sanger sequencing for methylation, suggesting it may serve as a potential noninvasive diagnostic tool for the detection of CRC. This clinical trial has been registered in ClinicalTrials.gov (NCT05508503).

Imaging Mutant Huntingtin Aggregates using [18F]CHDI-650 PET in a preclinical model of Huntington’s disease

Huntington’s disease (HD) is a fatal neurodegenerative disorder caused by a CAG repeat expansion in the huntingtin gene (HTT) that encodes pathogenic mutant huntingtin (mHTT) protein. Therapeutic strategies aimed at lowering mHTT would benefit from non-invasive PET-based methods to evaluate treatment response. The CHDI foundation developed the first-generation PET ligands labelled with 11C, which advanced to clinical evaluation. Further optimization generated an 18F radioligand with improved metabolic stability, brain exposure and specificity, shown by PET imaging in wild-type mice and autoradiography in HD brain samples1,2. Here, we characterised the PET radioligand [18F]CHDI-650 in the R6/1 mouse model of HD to assess its sensitivity for monitoring treatment response of mHTT lowering therapies. [18F]CHDI-650 was produced according to an established method3 using a TRACERlab system. Radiochemical purity/molar activity of the final [18F]CHDI-650 product was detected by reverse-phase HPLC analysis. Dynamic PET imaging was conducted using R6/1 mice and wild-type littermates at various disease stages. Image processing and kinetic modelling (2TCM) analysis were performed using PMOD software. Ex vivo brain radioactivity was measured using gamma counting. Brains were cryo-sectioned and processed for immunofluorescence to identify HTT load. [18F]CHDI-650 was produced (1.3 ± 0.2 GBq) with an average radiochemical purity of 99.9 ± 0.1 % and a molar activity of 32.4 ± 5.2 GBq/µmol (n = 9). The radioligand was able to discriminate R6/1 HET from WT mice at an early presymptomatic age for both dynamic PET imaging (striatal) and ex vivo whole brain gamma. A well-defined age associated increase was also evident using both measures. Dual immunofluorescence spatially defined the level of aggregated HTT in striatal and hippocampal regions across the age groups. [18F]CHDI-650 PET shows great potential as a non-invasive tool for visualising the development of mHTT pathology and as a biomarker for preclinical efficacy studies. Please click on the 'PDF' for the full abstract!

The Relationship between the Infrared Eye Temperature of Beef Cattle and Associated Biological Responses at High Environmental Temperatures.

Cattle in regions with high ambient temperatures are at risk of heat stress. Early detection is important to allow action to be taken to minimise the risks to cattle exposed to thermal stress. This study aimed to investigate the impact of heat stress on IRT-Eye temperature and its association with the behavioural and physiological responses of heat-stressed Angus steers (n = 24) on finisher and or substituted diets. Overall, 2 cohorts of 12 Angus steers were individually housed in a climate-controlled facility to examine responses to heat stress when fed on a standard finisher diet, based on a high percentage of cereal grains, and on a substituted diet in which 8% of the grains were replaced by an isoenergetic amount of lucerne hay. Exposing feedlot cattle to hot environmental conditions increased IRT-Eye temperature, which had a strong association with behaviour and physiology. There was no evidence of differences between the different dietary cohorts. The cattle with increased IRT-Eye temperature showed stress-related responses, including a downward-facing head, ears directed backwards, and other indicators of heat stress such as increased panting, standing, and increased rumen temperature. The strong association of IRT-Eye temperature with stress-related behaviours, as well as with rumen temperature and panting behaviour, highlights the potential for IRT-Eye to be utilised as a non-invasive tool to assess cattle responses in hot conditions.

Artificial intelligence-enhanced detection of subclinical coronary artery disease in athletes: diagnostic performance and limitations.

This study evaluates the diagnostic performance of artificial intelligence (AI)-based coronary computed tomography angiography (CCTA) for detecting coronary artery disease (CAD) and assessing fractional flow reserve (FFR) in asymptomatic male marathon runners. We prospectively recruited 100 asymptomatic male marathon runners over the age of 45 for CAD screening. CCTA was analyzed using AI models (CorEx and Spimed-AI) on a local server. The models focused on detecting significant CAD (≥ 50% diameter stenosis, CAD-RADS 3, 4, or 5) and distinguishing hemodynamically significant stenosis (FFR ≤ 0.8) from non-significant stenosis (FFR > 0.8). Statistical analysis included sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy. The AI model demonstrated high sensitivity, with 91.2% for any CAD and 100% for significant CAD, and high NPV, with 92.7% for any CAD and 100% for significant CAD. The diagnostic accuracy was 73.4% for any CAD and 90.4% for significant CAD. However, the PPV was lower, particularly for significant CAD (25.0%), indicating a higher incidence of false positives. AI-enhanced CCTA is a valuable non-invasive tool for detecting CAD in asymptomatic, low-risk populations. The AI model exhibited high sensitivity and NPV, particularly for identifying significant stenosis, reinforcing its potential role in screening. However, limitations such as a lower PPV and overestimation of disease indicate that further refinement of AI algorithms is needed to improve specificity. Despite these challenges, AI-based CCTA offers significant promise when integrated with clinical expertise, enhancing diagnostic accuracy and guiding patient management in low-risk groups.

Effect of closed-loop vibration stimulation on sleep quality for poor sleepers.

Recent studies have investigated the autonomic modulation method using closed-loop vibration stimulation (CLVS) as a novel strategy for enhancing sleep quality. This study aimed to explore the effects of CLVS on sleep quality, autonomic regulation, and brain activity in individuals with poor sleep quality. Twenty-seven participants with poor sleep quality (Pittsburgh sleep quality index >5) underwent two experimental sessions using polysomnography and a questionnaire, one with CLVS (STIM) and the other without (SHAM). Sleep macrostructure analysis first showed that CLVS significantly reduced the total time, proportion, and average duration of waking after sleep onset. These beneficial effects were paralleled by significantly increased self-reported sleep quality. Moreover, there was a significant increase in the normalized high-frequency (nHF) and electroencephalography relative powers of delta activity during N3 sleep under STIM. Additionally, coherence analysis between nHF and delta activity revealed strengthened coupling between cortical and cardiac oscillations. This study demonstrated that CLVS significantly improves sleep quality in individuals with poor sleep quality by enhancing both subjective and objective measures. These findings suggest that CLVS has the potential to be a practical, noninvasive tool for enhancing sleep quality in individuals with sleep disturbances, offering an effective alternative to pharmacological treatments.

Circulating tumour DNA in patients with hepatocellular carcinoma across tumour stages and treatments

ObjectiveCirculating tumour DNA (ctDNA) is a promising non-invasive biomarker in cancer. We aim to assess the dynamic of ctDNA in patients with hepatocellular carcinoma (HCC).DesignWe analysed 772 plasmas from 173...

Ranking attention multiple instance learning for lymph node metastasis prediction on multicenter cervical cancer MRI.

In the current clinical diagnostic process, the gold standard for lymph node metastasis (LNM) diagnosis is histopathological examination following surgical lymphadenectomy. Developing a non-invasive and preoperative method for predicting LNM is necessary and holds significant clinical importance. We develop a ranking attention multiple instance learning (RA-MIL) model that integrates convolutional neural networks (CNNs) and ranking attention pooling to diagnose LNM from T2 MRI. Our RA-MIL model applies the CNNs to derive imaging features from 2D MRI slices and employs ranking attention pooling to create patient-level feature representation for diagnostic classification. Based on the MIL and attention theory, informative regions of top-ranking MRI slices from LNM-positive patients are visualized to enhance the interpretability of automatic LNM prediction. This retrospective study collected 300 female patients with cervical cancer who underwent T2-weighted magnetic resonance imaging (MRI) scanning and histopathological diagnosis from one hospital (289 patients) and one open-source dataset (11 patients). Our RA-MIL model delivers promising LNM prediction performance, achieving the area under the receiver operating characteristic curve (AUC) of 0.809 on the internal test set and 0.833 on the public dataset. Experiments show significant improvements in LNM status prediction using the proposed RA-MIL model compared with other state-of-the-art (SOTA) comparative deep learning models. The developed RA-MIL model has the potential to serve as a non-invasive auxiliary tool for preoperative LNM prediction, offering visual interpretability regarding informative MRI slices and regions in LNM-positive patients.

State-of-the-Art in Skin Fluorescent Photography for Cosmetic and Skincare Research: From Molecular Spectra to AI Image Analysis.

Autofluorescence is a remarkable property of human skin. It can be excited by UV and observed in the dark using special detection systems. The method of fluorescence photography (FP) is an effective non-invasive tool for skin assessment. It involves image capturing by a camera the emission of light quanta from fluorophore molecules in the skin. It serves as a useful tool for cosmetic and skincare research, especially for the detection of pathological skin states, like acne, psoriasis, etc. To the best of our knowledge, there is currently no comprehensive review that fully describes the application and physical principles of FP over the past five years. The current review covers various aspects of the skin FP method from its biophysical basis and the main fluorescent molecules of the skin to its potential applications and the principles of FP recording and analysis. We pay particular attention to recently reported works on the automatic analysis of FP based on artificial intelligence (AI). Thus, we argue that FP is a rapidly evolving technology with a wide range of potential applications. We propose potential directions of the development of this method, including new AI algorithms for the analysis and expanding the range of applications.

Comparative study of microneedling monotherapy versus microneedling with autologous platelet-rich plasma for the treatment of stretch marks (striae distensae) and post-surgical scars: Clinical and dermoscopy outcomes

Objectives: Stretch marks and post-surgical scar marks may have negative cosmetic and psychological impacts.. Thus, this study aimed to compare the efficacy and safety of microneedling monotherapy versus microneedling in combination with autologous platelet-rich plasma (PRP) for the treatment of stretch marks and post-surgical scars. Material and Methods: This was a prospective randomized non-controlled study. A total of 30 study participants with stretch marks and post-surgery scars were enrolled and randomly divided into two groups, each with 15 study participants. All received three treatment sessions using microneedling monotherapy (Group A) or autologous PRP with microneedling (Group B) at 4-week intervals. Clinical response to treatment was assessed at the end of 20 weeks by comparing clinical photographs, Manchester Scar scale (MSS) score, dermoscopy images, and patient satisfaction scores. Results: At the end of the study, statistically significant esthetic improvement in terms of MSS scores was seen in Group B when compared to Group A. In Group A, 73.3% of the study participants were satisfied while in Group B, 86.6% were very satisfied with the results. Dermoscopy pictures substantiated the findings of clinical improvement. No major side effects were documented. Conclusion: PRP in combination with microneedling has a better outcome as compared to microneedling monotherapy. Furthermore, dermoscopy is a non-invasive tool that can be used to evaluate therapeutic response.

Versatile self-assembled near-infrared SERS nanoprobes for multidrug-resistant bacterial infection-specific surveillance and therapy

The rise of multidrug-resistant bacteria (MDRB) has made bacterial infection one of the biggest health threats, causing numerous antibiotics to fail. Real-time monitoring of bacterial disease treatment efficacy at the infection site is required. Herein, we report a versatile Raman tag 3,3′-diethylthiatricarbocyanine iodide (DTTC)-conjugated star-shaped Au-MoS2@hyaluronic acid (AMD@HA) nanocomposite as a surface-enhanced Raman scattering (SERS) nanoprobe for quick bacterial identification and in-situ eradication. Localized surface plasmon resonance (LSPR) from the hybrid metallic nanostructure makes AMD@HA highly responsive to the near-infrared laser, enabling it to demonstrate a photothermal (PTT) effect, increased SERS activity, and peroxidase-like catalytic reaction to release reactive oxygen species. The tail vein injection of AMD@HA nanoprobes is invasive, however SERS imaging for bacterial identification is non-invasive and sensitive, making it an efficient residual bacteria monitoring method. The detection limit for methicillin-resistant Staphylococcus aureus (MRSA) is as low as 102 CFU·mL-1, and the substrates allow for taking 120 s to acquire a Raman image of 1600 (40 × 40) pixels. In mouse models of MRSA-induced wound infection and skin abscess, the combination of AMD@HA-mediated PTT and catalytic therapy demonstrates a synergistic effect in promoting wound healing through rapid sterilization. This SERS-guided therapeutic approach exhibits little toxicity and does not cause considerable collateral damage, offering a highly promising intervention for treating diseases caused by MDRB. Statement of significanceThis research introduces a SERS nanoprobe, AMD@HA, for the rapid identification and eradication of multidrug-resistant bacteria (MDRB), a critical health threat. The nanoprobe leverages localized surface plasmon resonance for photothermal therapy and enhanced Raman signals, offering a sensitive, non-invasive diagnostic tool. With a low detection limit for MRSA and a synergistic therapeutic effect in mouse models, our approach holds significant promise for treating MDRB-driven infections with minimal toxicity, advancing the field of antimicrobial strategies.

12591 Screening For Metabolic Dysfunction-Associated Steatotic Liver Disease In Diabetes Clinics

Abstract Disclosure: D. D'Annibale: None. M. Mimoto: None. R. Lis: None. A. Deioma: None. H. Patton: None. Screening for Metabolic Dysfunction-Associated Steatotic Liver Disease in Diabetes Clinics Background: The prevalence of metabolic dysfunction-associated steatotic liver disease (MASLD) in patients with Type 1 (T1D) or Type 2 diabetes (T2D) is elevated, ranging from 22% in T1D to >55% in T2D. However, MASLD screening is not standard in diabetes clinics. The Fibrosis-4 (FIB-4) score and transient elastography (TE) are non-invasive tools to identify those at risk for MASLD and liver fibrosis. To identify at-risk individuals and determine the prevalence of undiagnosed MASLD in our Veterans with diabetes and elevated BMI, we developed a screening protocol using FIB-4 and TE. Methods: Veterans with T1D or T2D and elevated BMI seen in diabetes clinic were prospectively screened. Patients with prior TE or liver disease were excluded. In phase 1, patients were offered in-clinic TE during their visit. In phase 2, a FIB-4 determined referral to TE or hepatology. Results: 274 patients were screened over 8 months. Of 102 qualifying patients in phase 1, 28 had in-clinic TE; of these, 7 (25%) had a kPa of 8 or greater, indicating increased risk of fibrosis, and 16 (57.1%) had a CAP score >280, indicating severe steatosis.In phase 2, a FIB-4 was ordered in 157 qualifying patients. Of 110 completed tests, the FIB-4 was <1.3 (low risk), 1.3–2.6 (indeterminate risk) and >2.6 (high risk) in 61 (55.5%), 44 (40%) and 5 (4.5%), respectively. Patients were referred to TE (indeterminate) or hepatology (high risk). To date, 21 patients have undergone TE, 5 (23.8%) had a kPa of 8 or greater and 11 (52.4%) had a CAP score >280. Conclusions In Veterans with T1D or T2D and elevated BMI, both of our screening protocols identified those with severe steatosis (57.1% vs 52.4%) by TE at rates similar to those reported in the literature. Increased risk of fibrosis by TE was identified in ∼24-25% of our screened population. The high prevalence of undiagnosed MASLD in our Veteran population supports the need for screening in diabetes clinics. Increased provider awareness and point-of-care TE may increase access to screening and improve early diagnosis. Presentation: 6/2/2024

Deep Learning Models for Early Detection of Ovarian Cancer: A Systematic Review of Ultrasound-Based Diagnostic Tools

Ovarian Cancer (OC) remains one of the most lethal gynecological malignancies, with a global 5-year survival rate of only 45%. This project aimed to assess the potential of deep learning-based diagnostic tools in improving early detection of ovarian cancer, particularly through ultrasound imaging. The primary research problem lies in the challenge of accurately diagnosing OC at an early stage, as traditional imaging techniques often rely on subjective interpretations, leading to inconsistent results. To address this issue, a systematic review was conducted following PRISMA guidelines, evaluating studies published between January 2018 and May 2023 in the Scopus and PubMed databases. These studies employed deep learning or artificial intelligence models for the diagnosis, prognosis, or identification of OC from ultrasound images. Of the 101 studies screened, 9 met the inclusion criteria. The included studies reported diagnostic accuracies of deep learning models ranging from 75% to 100%, with sensitivities between 85% and 99%. The conclusions indicate that deep learning models significantly enhance the diagnostic accuracy of ovarian cancer, offering a promising non-invasive tool for early detection. This research underscores the importance of integrating AI technologies into clinical practice to improve survival outcomes for OC patients.