Accuracy For Detection Of Lesions Research Articles

Leveraging Pre-Trained Transformers for Efficient Segmentation and Lesion Detection in Cone-Beam CT Scans

IntroductionCone-beam computed tomography (CBCT) is widely used to detect jaw lesions, although CBCT interpretation is time-consuming and challenging. Artificial intelligence (AI) for CBCT segmentation may improve lesion detection accuracy. However, consistent automated lesion detection remains difficult, especially with limited training data. This study aimed to assess the applicability of pre-trained transformer-based architectures for semantic segmentation of CBCT volumes when applied to periapical lesion detection. MethodsCBCT volumes (n=138) were collected and annotated by expert clinicians using five labels – "lesion", "restorative material", "bone", "tooth structure", and "background". U-Net (convolutional neural network (CNN)-based) and Swin-UNETR (transformer-based) models, pre-trained (Swin-UNETR-PRETRAIN) and from scratch (Swin-UNETR-SCRATCH), were trained with subsets of the annotated CBCTs. These models were then evaluated for semantic segmentation performance using the Sørensen–Dice coefficient (DICE), lesion detection performance using sensitivity and specificity, and training sample size requirements by comparing models trained with 20, 40, 60, or 103 samples. ResultsTrained with 103 samples, Swin-UNETR-PRETRAIN achieved a DICE of 0.8512 for "lesion", 0.8282 for "restorative materials", 0.9178 for "bone", 0.9029 for "tooth structure", and 0.9901 for "background". “Lesion” DICE was statistically similar between Swin-UNETR-PRETRAIN trained with 103 and 60 images (P>.05), with the latter achieving 1.00 sensitivity and 0.94 specificity in lesion detection. With small training sets, Swin-UNETR-PRETRAIN outperformed Swin-UNETR-SCRATCH in DICE over all labels (P<.001 [n=20], P<.001 [n=40]), and U-Net in lesion detection specificity (P=.006 [n=20], P=.031 [n=40]). ConclusionsTransformer-based Swin-UNETR architectures allowed for excellent semantic segmentation and periapical lesion detection. Pre-trained, it may provide an alternative with smaller training datasets compared to classic U-Net architectures.

RefineDet with Improved Attention Block for Chest X-ray Image Lesion Detection

In recent years, artificial intelligence technology has been widely used to assist radiologists in diagnosing and analyzing medical images. The use of artificial intelligence technology can well assist doctors in the localization of lesions. However, the mainstream target detection models at this stage are difficult to be practically applied in medical systems because of factors such as the use of large backbone networks and high input resolution, which leads to low model accuracy and high consumption of computational resources. In this paper, we propose a fast detection speed and high accuracy of the lung lesion detection network IAB- RefineDet. By improving the channel and spatial attention mechanisms and introducing the improved attention module into RefineDet, the lesion detection accuracy is dramatically improved without significantly increasing the number of parameters. We conduct extensive experiments on VinDr-CXR, the world's largest publicly available chest radiograph detection dataset, and comparative experiments with existing mainstream target detection models. The experimental results show that IAB-RefineDet achieves a mAP of 16.23%, and the lesion detection performance is significantly better than mainstream deep learning models.

A SWOT ANALYSIS OF BREAST CANCER DIAGNOSIS IN DIGITAL MAMMOGRAPHY USING DEEP CONVOLUTIONAL NEURAL NETWORK

Breast cancer is a significant global health challenge that affects both men and women, leading to cause-specific deaths. Current early screening interventions, such as digital mammography (DM), are susceptible to high false-positives and false-negatives. This paper explores the potential of convolutional neural network (CNN), a form of artificial intelligence (AI), to support screening mammography with the aim to enhance accuracy in lesion detection, image classification and diagnostic prediction. Because the adoption of AI in cancer diagnosis is still in its infancy, the objective of this paper is to provide insight into the benefits and limitations of deep learning-based approaches to detect and diagnose cancer. An analysis of the implementation of CNN in AI-screening mammography models was conducted, using the SWOT strategic analysis tool. Internal strengths that improve the predictive accuracy of CNN include transfer learning and data augmentation, whereas the internal weaknesses include a lack of data standardisation and reproducibility. External opportunities consist of increased sensitivity in differentiating between microcalcifications and non-tumorous structures, improved predictive diagnosis and reduced workload. Nevertheless, integration within clinical settings must also consider the external threats of breaching patient privacy, automation biases and the role of clinical judgement.

Breast Cancer and Lesion Detection using Machine Learning : A Review

Breast Breast cancer poses a significant global health concern, with approximately 2.2 million new cases and 700,000 deaths reported in 2020. Traditional diagnostic approaches which predominantly depend on expert judgement, have been associated with substantial variability in accuracy. To bridge this gap ML models are used to improve diagnostic out of which the present research investigates the potential of specific machine learning algorithms—D cancer remains one of the most common cancers among women globally, necessitating early detection to improve prognosis and survival rates. Recent advancements in machine learning (ML) have shown promise in enhancing the accuracy and efficiency of breast cancer and lesion detection. This review paper discusses the current methodologies, challenges, and future directions of ML applications in the detection and diagnosis of breast cancer. We analyze various ML algorithms used for analyzing mammograms, ultrasound, and MRI data, evaluating their effectiveness and applicability in clinical settings.

Crow search algorithm with deep transfer learning driven skin lesion detection on dermoscopic images

The early diagnosis of skin cancer is of paramount importance for effective patient treatment. Dermoscopy, a non-surgical technique, utilizes precise equipment to examine the skin and plays a crucial role in identifying specific features and patterns that may indicate the presence of skin carcinoma. In recent times, machine learning (ML) methods have been developed to recognize and classify dermoscopic images as either malignant or benign. Deep learning (DL) systems, including Convolutional Neural Networks (CNNs), as well as various ML models like Random Forest (RF) classifiers and Support Vector Machine (SVM), are employed to extract relevant features from these images. This study introduces the Crow Search Algorithm with Deep Transfer Learning Driven Skin Lesion Detection on Dermoscopic Images (CSADTL-SLD) technique. The CSADTL-SLD method starts with the application of a median filter (MF) to remove noise from the images and utilizes the GoogleNet model for feature extraction. GoogleNet is well-regarded for its capacity to capture intricate and meaningful patterns within the data, which are essential for accurate lesion characterization. Furthermore, the CSADTL-SLD technique applies the Crow Search Algorithm (CSA) for parameter tuning of the GoogleNet model. After feature selection, the system employs the MLP classification model for precise lesion categorization. The comprehensive results of this research demonstrate the superiority of the CSADTL-SLD algorithm, showing significant enhancements in skin lesion detection accuracy and robustness when compared to existing methods. This approach holds promise as an effective solution for automating the detection and classification of skin lesions in dermoscopic images.

Phantom study and clinical application of total-body 18F-FDG PET/CT imaging: How to use small voxel imaging better?

BackgroundConventional PET/CT imaging reconstruction is typically performed using voxel size of 3.0–4.0 mm in three axes. It is hypothesized that a smaller voxel sizes could improve the accuracy of small lesion detection. This study aims to explore the advantages and conditions of small voxel imaging on clinical application.MethodsBoth NEMA IQ phantom and 30 patients with an injected dose of 3.7 MBq/kg were scanned using a total-body PET/CT (uEXPLORER). Images were reconstructed using matrices of 192 × 192, 512 × 512, and 1024 × 1024 with scanning duration of 3 min, 5 min, 8 min, and 10 min, respectively.ResultsIn the phantom study, the contrast recovery coefficient reached the maximum in matrix group of 512 × 512, and background variability increased as voxel size decreased. In the clinical study, SUVmax, SD, and TLR increased, while SNR decreased as the voxel size decreased. When the scanning duration increased, SNR increased, while SUVmax, SD, and TLR decreased. The SUVmean was more reluctant to the changes in imaging matrix and scanning duration. The mean subjective scores for all 512 × 512 groups and 1024 × 1024 groups (scanning duration ≥ 8 min) were over three points. One false-positive lesion was found in groups of 512 × 512 with scanning duration of 3 min, 1024 × 1024 with 3 min and 5 min, respectively. Meanwhile, the false-negative lesions found in group of 192 × 192 with duration of 3 min and 5 min, 512 × 512 with 3 min and 1024 × 1024 with 3 min and 5 min were 5, 4, 1, 4, and 1, respectively. The reconstruction time and storage space occupation were significantly increased as the imaging matrix increased.ConclusionsPET/CT imaging with smaller voxel can improve SUVmax and TLR of lesions, which is advantageous for the diagnosis of small or hypometabolic lesions if with sufficient counts. With an 18F-FDG injection dose of 3.7 MBq/kg, uEXPLORER PET/CT imaging using matrix of 512 × 512 with 5 min or 1024 × 1024 with 8 min can meet the image requirements for clinical use.

Improving small sample medical image segmentation using CBAM: Insights from two datasets

Medical image segmentation is one of the key tasks in the medical field and is crucial for accurate lesion detection and treatment planning. However, the small sample problem has been one of the challenges in medical image segmentation. In this study, the small sample medical image segmentation problem is evaluated experimentally based on two different datasets, the VOC dataset and the esophageal medical images. In the VOC dataset experiments, we used the code from the BAM project on GitHub as a benchmark and compared it. Additionally, we improved the benchmark code by adding the Attention Mechanism (CBAM), placing its position before the support set undergoes global average pooling. By comparing the experimental results, we found that the model with the added CBAM achieved better segmentation results under small sample conditions. In the esophageal medical image experiments, the code from the BAM project was also used as a benchmark and a similar experimental design was performed. We further added CBAM by placing it before the global average pooling of the support set. The experimental results show that the model with the addition of CBAM achieves a significant improvement in the small sample segmentation task of esophageal medical images. To summarize, this study validates the effectiveness of using the BAM model in the small-sample medical image segmentation task on two different datasets. By adding CBAM, our model exhibits better segmentation performance under small sample conditions. These findings provide useful insights for small-sample medical image segmentation and are important for future research and practice.

MSPA-DLA++: A Multi-Scale Phase Attention Deep Layer Aggregation for Lesion Detection in Multi-Phase CT Images.

Object detection using convolutional neural networks (CNNs) has achieved high performance and achieved state-of-the-art results with natural images. Compared to natural images, medical images present several challenges for lesion detection. First, the sizes of lesions vary tremendously, from several millimeters to several centimeters. Scale variations significantly affect lesion detection accuracy, especially for the detection of small lesions. Moreover, the effective extraction of temporal and spatial features from multi-phase CT images is also an important issue. In this paper, we propose a group-based deep layer aggregation method with multiphase attention for liver lesion detection in multi-phase CT images. The method, which is called MSPA-DLA++, is a backbone feature extraction network for anchor-free liver lesion detection in multi-phase CT images that addresses scale variations and extracts hidden features from such images. The effectiveness of the proposed method is demonstrated on public datasets (LiTS2017) and our private multiphase dataset. The results of the experiments show that MSPA-DLA++ can improve upon the performance of state-of-the-art networks by approximately 3.7%.

A polymeric 1H/19F dual-modal MRI contrast agent with a snowman-like Janus nanostructure.

Magnetic resonance imaging (MRI) has emerged as a pivotal tool in contemporary medical diagnostics, offering non-invasive and high-resolution visualization of internal structures. Contrast agents are essential for enhancing MRI resolution, accurate lesion detection, and early pathology identification. While gadolinium-based contrast agents are widely used in clinics, safety concerns have prompted exploration of metal-free alternatives, including fluorine and nitroxide radical-based MRI contrast agents. Fluorine-containing compounds exhibit excellent MRI capabilities, with 19F MRI providing enhanced resolution and quantitative assessment. Nitroxide radicals, such as PROXYL and TEMPO, offer paramagnetic properties for MRI contrast. Despite their versatility, nitroxide radicals suffer from lower relaxivity values (r1) compared to gadolinium. Dual-modal imaging, combining 1H and 19F MRI, has gained prominence for its comprehensive insights into biological processes and disease states. However, existing dual-modal agents predominantly utilize gadolinium-organic ligands without incorporating nitroxide radicals. Here, we introduce a novel dual-modal MRI contrast agent (J-CA) featuring a Janus asymmetric nanostructure synthesized via seeded emulsion polymerization and post-modification. J-CA demonstrates excellent in vitro and in vivo performance in both 19F and 1H MRI, with a T2 relaxation time of 5 ms and an r1 value of 0.31 mM-1 s-1, ensuring dual-modal imaging capability. Moreover, J-CA exhibits superior biocompatibility and organ targeting, making it a promising candidate for precise lesion imaging and disease diagnosis. This work introduces a new avenue for metal-free dual-modal MRI, addressing safety concerns associated with traditional contrast agents.

Potential Usefulness a Coronal View using an Automated Breast Ultrasound System in Detecting Breast Lesions.

An automated breast ultrasound system (ABUS) combined with screening mammography has increased cancer detection rates; however, supplemental ABUS use has increased recall rates. In this study, we aimed to identify an accurate and efficient method of ABUS interpretation and evaluate the potential usefulness of its coronal view versus the conventional transverse view. This retrospective observer study included comprised 114 ABUS cases (40 normal, 35 benign, 39 malignant). Ten physicians from multiple institutions interpreted the anonymized coronal and transverse views independently. The observers scored their confidence in the lesion detection for each case using a continuous scale and recorded reading times for each coronal and transverse view interpretation. Free-response receiver operating characteristic analysis was employed to compare detection accuracies between views; a paired t-test was used to compare the average reading times. Detection accuracy did not differ significantly between the coronal and transverse views (figure of merit=0.740 and 0.745, respectively; p = 0.72). However, the average reading time for the coronal view was significantly shorter than that for the transverse view (149.7 vs. 200.3 seconds per case, p = 0.003). The coronal view obtained with the ABUS was useful for interpretation and associated with significantly shorter reading times compared with the conventional transverse view while maintaining breast lesion detection accuracy.

SC-UneXt: Nested UNeXt Architecture based on Medical Image Segmentation

UNet and its various variants are commonly used methods in medical image segmentation tasks; however, many network parameters, complex calculations, and slow usage are problems that need to be overcome. These problems hinder the specific application of fast image segmentation in real-time tasks. At the same time, the lesion area has problems such as small size, irregular shape, and blurred edges, which makes the network feature extraction difficult and the segmentation accuracy needs to be improved. At the same time, medical image segmentation provides a variety of effective methods for the accuracy and robustness of organ segmentation, lesion detection, and classification. Medical images have fixed structures, simple semantics, and diverse details, so integrating rich multi-scale features can improve segmentation accuracy. Given that the density of diseased tissue may be comparable to that of surrounding normal tissue, both global and local information are crucial to segmentation results. To this end, we propose an image segmentation method (SC -UNe X t) based on edge feature extraction and multi-scale feature fusion of convolutional multi-layer perceptron (MLP). The network is a deeply supervised encoder-decoder network, in which the encoder and decoder pass through a series of nested, multiple jump paths to reduce the semantic gap between the feature maps of the encoder and decoder sub-networks.; Multi - scale feature fusion is introduced based on the UNe Finally, we evaluate our model approach on the LIDC dataset public dataset. Experiments have proven the effectiveness of this method. Our model's similarity coefficient and intersection ratio reached 86.44% and 90.86% respectively. Compared with UNet and UNe X t, the network proposed in this article has improved in accuracy, intersection ratio of real values and predicted values, similarity coefficient, and segmentation effect.

Chromocolposcopy with fluorescein sodium for detection of premalignant and malignant lesions of the cervix: A cross-sectional study.

The aim of the study was to explore the utility of fluorescein sodium (FNa) as a contrast agent for colposcopy to detect premalignant and malignant lesions of cervix. The primary objective was to determine and compare the percentage detection of premalignant and malignant lesions of FNa and acetic acid (AA) positive areas. This study included 120 screen positive women who underwent colposcopy using both 3% AA and FNa (0.06%). Observations for FNa staining were made under blue filter and directed biopsies were taken from acetowhite and fluorescent green areas. Benign lesions were considered as disease-negative and low grade squamous intraepithelial lesions (LSIL), high grade SIL (HSIL), and invasive cancer were considered as disease-positive. Correlation between histopathology and FNa and AA was determined by Kappa statistics. The mean age was 39.59 ± 10.73 years and median parity was 2. Out of 120 patients, 57 had benign lesions, 18 had LSIL, 33 had HSIL and 12 had invasive carcinomas. Sensitivity was 98.41% versus 64.91% respectively and specificity was 85.71% versus 35.09% respectively with FNa and AA. Diagnostic accuracy of FNa and AA was 82.50% versus 61.60%. There was good agreement between FNa staining and final histopathology and fair agreement between AA application and HPE (κ = 0.643 vs 0.213, P < 0.001). Using FNa as a contrast agent during colposcopy results in greater accuracy for detection of premalignant and malignant lesions of the cervix as compared to AA.

Comparison of conventional MR arthrography and 3D volumetric MR arthrography in detection of cartilage defects accompanying glenoid labrum pathologies.

In this study, we aimed to compare conventional and T1-weighted volumetric magnetic resonance arthrography (MRA) in the diagnosis and grading of glenoid cartilage defects that accompany labral pathologies. A total of 79 patients who were prediagnosed with labrum pathologies based on shoulder magnetic resonance imaging (MRI) had MRA and CTA between December 2021 and May 2022. CTA was regarded as reference standard. CTA images were examined by a radiologist experienced in musculoskeletal radiology, and MRA images were examined by two radiologists independently to determine presence, grade, and localization of any glenoid cartilage defect, if present. Sensitivity, specificity, and accuracy were calculated separately for conventional and T1-weighted volumetric MRA. In addition, at the last stage, two observers examined all MRAs together, and the presence of a cartilage defect was decided by consensus, and the overall sensitivity, specificity, and accuracy were calculated. Cartilage defect was detected on CTAs of 48 (60.75%) cases of among 79 patients with labrum pathology. The sensitivity, specificity, and accuracy of conventional MRA for two examiners were 17-19%, 100-100%, and 49-51%, respectively, while those values were 67-65%, 92-97%, and 84-77%, respectively, for T1-weighted volumetric MRA. Inter-examiner agreement was excellent for diagnosis of cartilage defects on all MRAs. The overall sensitivity, specificity, and accuracy for detection of glenoid cartilage lesions by MRA were 69%, 97%, and 80%, respectively. T1-weighted volumetric MRA seems to demonstrate cartilage defects accompanied with labrum pathologies accurately with high sensitivity, specificity, and excellent inter-examiner agreement.

The value of a deep learning image reconstruction algorithm in whole-brain computed tomography perfusion in patients with acute ischemic stroke.

Computed tomography perfusion (CTP) and computed tomography angiography (CTA) are valuable tools for diagnosing acute ischemic stroke (AIS). It is essential to obtain high-quality CTP and CTA images in short time. This study aimed to evaluate the image quality and diagnostic performance of brain CTP and CTA images generated from CTP reconstructed by a deep learning image reconstruction (DLIR) algorithm on patients with AIS. The study prospectively enrolled 54 patients with suspected AIS undergoing non-contrast CT and CTP within 24 hours. CTP datasets were reconstructed with three levels of adaptive statistical iterative reconstruction-Veo algorithm [ASIR-V 0% with filtered back projection (FBP), ASIR-V 40%, and ASIR-V 80%] and three levels of DLIR, including low (DLIR-L), medium (DLIR-M), and high (DLIR-H). CTA images were generated using the CTP datasets at the peak arterial phase. Objective parameters including signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and noise reduction rate. Subjective evaluation was assessed according to Abels scoring system. Perfusion parameters and detection accuracy for infarction core lesions were evaluated. The objective and subjective image quality of CTA images were also evaluated. All reconstructions produced similar CT values (P>0.05). With the increase of ASIR-V and DLIR reconstruction strength, image noise decreased, while SNR and CNR increased for CTP images, especially in white matter. DLIR-H, DLIR-M, and ASIR-V80% yielded higher subjective scores than did ASIR-V40% and FBP. DLIR-H provided the highest noise reduction rate and detection accuracy. No significant difference was found in conventional parameters, the volume of infarct core, or ischemic penumbra among the 6 groups (P>0.05). The objective evaluation of reconstructed CTA images was comparable in DLIR-H, DLIR-M, and ASIR-V80% (P>0.05). The subjective scores of the DLIR-H and DLIR-M images were higher than those of the other groups, especially ASIR-V40% and FBP (P<0.05). Compared with FBP and ASIR-V40%, DLIR-H, DLIR-M, and ASIR-V80% improved the overall image quality of CTP and CTA images to varying degrees. Furthermore, DLIR-H and DLIR-M showed the best performance. DLIR-H is the best choice in diagnosing AIS with improved detection accuracy for cerebral infarction. Reconstructing CTA images using CTP datasets could reduce contrast agent and radiation dose.

Uncertainty-based Active Learning by Bayesian U-Net for Multi-label Cone-beam CT Segmentation

IntroductionTraining of Artificial Intelligence (AI) for biomedical image analysis depends on large annotated datasets. This study assessed the efficacy of Active Learning (AL) strategies training AI models for accurate multilabel segmentation and detection of periapical lesions in cone-beam CTs (CBCTs) using a limited dataset. MethodsLimited field-of-view CBCT volumes (n = 20) were segmented by clinicians (clinician segmentation [CS]) and Bayesian U-Net-based AL strategies. Two AL functions, Bayesian Active Learning by Disagreement [BALD] and Max_Entropy [ME], were used for multilabel segmentation (“Lesion”-“Tooth Structure”-“Bone”-“Restorative Materials”-“Background”), and compared to a non-AL benchmark Bayesian U-Net function. The training-to-testing set ratio was 4:1. Comparisons between the AL and Bayesian U-Net functions versus CS were made by evaluating the segmentation accuracy with the Dice indices and lesion detection accuracy. The Kruskal-Wallis test was used to assess statistically significant differences. ResultsThe final training set contained 26 images. After 8 AL iterations, lesion detection sensitivity was 84.0% for BALD, 76.0% for ME, and 32.0% for Bayesian U-Net, which was significantly different (P < .0001; H = 16.989). The mean Dice index for all labels was 0.680 ± 0.155 for Bayesian U-Net and 0.703 ± 0.166 for ME after eight AL iterations, compared to 0.601 ± 0.267 for Bayesian U-Net over the mean of all iterations. The Dice index for “Lesion” was 0.504 for BALD and 0.501 for ME after 8 AL iterations, and at a maximum 0.288 for Bayesian U-Net. ConclusionsBoth AL strategies based on uncertainty quantification from Bayesian U-Net BALD, and ME, provided improved segmentation and lesion detection accuracy for CBCTs. AL may contribute to reducing extensive labeling needs for training AI algorithms for biomedical image analysis in dentistry.

The Role of Artificial Intelligence in Colorectal Cancer Screening: Lesion Detection and Lesion Characterization

Colorectal cancer remains a leading cause of cancer-related morbidity and mortality worldwide, despite the widespread uptake of population surveillance strategies. This is in part due to the persistent development of ‘interval colorectal cancers’, where patients develop colorectal cancer despite appropriate surveillance intervals, implying pre-malignant polyps were not resected at a prior colonoscopy. Multiple techniques have been developed to improve the sensitivity and accuracy of lesion detection and characterisation in an effort to improve the efficacy of colorectal cancer screening, thereby reducing the incidence of interval colorectal cancers. This article presents a comprehensive review of the transformative role of artificial intelligence (AI), which has recently emerged as one such solution for improving the quality of screening and surveillance colonoscopy. Firstly, AI-driven algorithms demonstrate remarkable potential in addressing the challenge of overlooked polyps, particularly polyp subtypes infamous for escaping human detection because of their inconspicuous appearance. Secondly, AI empowers gastroenterologists without exhaustive training in advanced mucosal imaging to characterise polyps with accuracy similar to that of expert interventionalists, reducing the dependence on pathologic evaluation and guiding appropriate resection techniques or referrals for more complex resections. AI in colonoscopy holds the potential to advance the detection and characterisation of polyps, addressing current limitations and improving patient outcomes. The integration of AI technologies into routine colonoscopy represents a promising step towards more effective colorectal cancer screening and prevention.

Video Analysis of Small Bowel Capsule Endoscopy Using a Transformer Network.

Although wireless capsule endoscopy (WCE) detects small bowel diseases effectively, it has some limitations. For example, the reading process can be time consuming due to the numerous images generated per case and the lesion detection accuracy may rely on the operators' skills and experiences. Hence, many researchers have recently developed deep-learning-based methods to address these limitations. However, they tend to select only a portion of the images from a given WCE video and analyze each image individually. In this study, we note that more information can be extracted from the unused frames and the temporal relations of sequential frames. Specifically, to increase the accuracy of lesion detection without depending on experts' frame selection skills, we suggest using whole video frames as the input to the deep learning system. Thus, we propose a new Transformer-architecture-based neural encoder that takes the entire video as the input, exploiting the power of the Transformer architecture to extract long-term global correlation within and between the input frames. Subsequently, we can capture the temporal context of the input frames and the attentional features within a frame. Tests on benchmark datasets of four WCE videos showed 95.1% sensitivity and 83.4% specificity. These results may significantly advance automated lesion detection techniques for WCE images.

Lung Nodule Detection With Modern Low-Field MRI (0.55 T) in Comparison to CT.

The aim of this study was to evaluate the accuracy of modern low-field magnetic resonance imaging (MRI) for lung nodule detection and to correlate nodule size measurement with computed tomography (CT) as reference. Between November 2020 and July 2021, a prospective clinical trial using low-field MRI at 0.55 T was performed in patients with known pulmonary nodules from a single academic medical center. Every patient underwent MRI and CT imaging on the same day. The primary aim was to evaluate the detection accuracy of pulmonary nodules using MRI with transversal periodically rotated overlapping parallel lines with enhanced reconstruction in combination with coronal half-Fourier acquired single-shot turbo spin-echo MRI sequences. The secondary outcome was the correlation of the mean lung nodule diameter with CT as reference according to the Lung Imaging Reporting and Data System. Nonparametric Mann-Whitney U test, Spearman rank correlation coefficient, and Bland-Altman analysis were applied to analyze the results. A total of 46 participants (mean age ± SD, 66 ± 11 years; 26 women) were included. In a blinded analysis of 964 lung nodules, the detection accuracy was 100% for those ≥6 mm (126/126), 80% (159/200) for those ≥4-<6 mm, and 23% (147/638) for those <4 mm in MRI compared with reference CT. Spearman correlation coefficient of MRI and CT size measurement was r = 0.87 ( P < 0.001), and the mean difference was 0.16 ± 0.9 mm. Modern low-field MRI shows excellent accuracy in lesion detection for lung nodules ≥6 mm and a very strong correlation with CT imaging for size measurement, but could not compete with CT in the detection of small nodules.

Utilizing Deep Learning for X-ray Imaging: Detecting and Classifying Degenerative Spinal Conditions.

Background Degenerative spinal conditions (DSCs) involve a diverse set of pathologies that significantly impact health and quality of life, affecting many individuals at least once during their lifetime. Treatment approaches are varied and complex, reflecting the intricacy of spinal anatomy and kinetics. Diagnosis and management pose challenges, with the accurate detection of lesions further complicated by age-related degeneration and surgical implants. Technological advancements, particularly in artificial intelligence (AI) and deep learning, have demonstrated the potential to enhance detection of spinal lesions. Despite challenges in dataset creation and integration into clinical settings, further research holds promise forimproved patient outcomes. Methods This study aimed to develop a DSCdetection and classification model using a Kaggle dataset of 967 spinal X-ray images at the Department of Neurosurgery of Arrowhead Regional Medical Center, Colton, California, USA.Our entire workflow, including data preprocessing, training, validation, and testing, was performed by utilizing an online-cloud based AI platform. The model's performance was evaluated based on its ability to accurately classify certain DSCs (osteophytes, spinal implants, and foraminal stenosis) and distinguish these from normal X-rays. Evaluation metrics, including accuracy, precision, recall, and confusion matrix, were calculated. Results The model achieved an average precision of 0.88, with precision and recall values of 87% and 83.3%, respectively, indicating its high accuracy in classifying DSCs and distinguishing these from normal cases. Sensitivity and specificity values were calculated as 94.12% and 96.68%, respectively. The overall accuracy of the model was calculated to be 89%. Conclusion These findings indicate the utility of deep learning algorithms in enhancing early DSC detection and screening. Our platform is a cost-effective tool that demonstrates robust performance given a heterogeneous dataset. However, additional validation studies are required to evaluate the model's generalizability across different populations and optimize its seamless integration into various types of clinical practice.

Added value of zoomed-echo-planar imaging diffusion-weighted imaging for evaluation of periampullary carcinomas.

To assess the image quality feasibility and diagnostic value of zoomed diffusion-weighted imaging (z-EPI DWI) using echo-planar imaging (EPI) compared with conventional DWI (c-EPI DWI) in patients with periampullary disease. Thirty-six patients with periampullary carcinomas and fifteen with benign periampullary disease were included in this study. All the subjects underwent MR cholangiopancreatography (MRCP), c-EPI DWI, and z-EPI DWI. Two radiologists independently assessed image quality of the two image sets, including overall image quality and lesion conspicuity. In addition, signal intensity and ADC measurements of DWIs in the periampullary lesions were conducted. Diagnostic accuracies of the combined image sets of MRCP and z-EPI DWI were compared with those of a combined set of MRCP and c-EPI DWI. z-EPI DWI showed significantly better image quality scores (anatomic structure visualization, 2.94 ± 0.24; overall image quality, 2.96 ± 0.17) compared to that with c-EPI DWI (anatomic structure visualization, 2.02 ± 0.22; overall image quality, 2.04 ± 0.24) (both P < 0.01). For all the periampullary malignant lesions and small lesions (≤20 mm), there was better delineation of lesion conspicuity and the lesion margin, as well as diagnostic confidence with z-EPI DWI (all P<0.05). The rate of periampullary malignancy's hyperintense signal on z-EPI DWI was increased to 91.7% (33/36) compared to c-EPI DWI (69.4% (25/36)) (P = 0.023). For all malignant lesions and small lesions, the diagnostic accuracy scores were increased using the MRCP and z-EPI DWI combined set, compared to the MRCP and c-EPI DWI combined set (P<0.05). Diagnostic accuracy for detection and differentiation of malignant lesions from benign lesions significantly improved for the MRCP and z-EPI DWI combined set compared with MRCP and c-EPI DWI combined set (P<0.05). There were no significant differences between c-EPI DWI and z-EPI DWI in the ADC values of periampullary malignant and benign lesions (P>0.05). z-EPI DWI has an advantage that could lead to remarkable image quality improvements and enhanced lesion visualization of periampullary carcinomas. z-EPI DWI was superior to c-EPI DWI for detecting, delineating, and diagnosing the lesions, particularly for small challenging lesions.