Detection Of Abnormalities Research Articles

8254 The Risk Of Thrombotic Complications In Patients With Post-COVID Syndrome Is High Even With Normal Coagulogram Values

Abstract Disclosure: A. Alieva: None. The high risk of thrombotic complications in patients with diabetes, even more increased in the post-Covid period, requires practicing physicians to be especially vigilant and carefully monitor the state of the hemostatic system and the function of the vascular endothelium. Routine laboratory tests may be insufficient for dynamic monitoring and timely detection of abnormalities. In this article, the author proposes the determination of cell adhesion molecules as a predictor of thrombotic complications in patients with type 2 diabetes mellitus in the post-Covid period. In our current study, VCAM-1 levels were elevated in all patients with type 2 diabetes within 24 months of COVID-19 infection and were not associated with increased von Willebrand factor or D-dimer or other coagulation abnormalities. This means that routine determination of coagulogram, D-dimer and von Willebrand factor may miss the stage of prothrombotic changes in the microvasculature. It is possible that an increase in the level of cell adhesion molecules precedes the appearance of other abnormalities in the hemostasiogram. Therefore, for patients with type 2 diabetes after a coronavirus infection with normal coagulogram values, it seems promising to determine the level of VCAM-1 molecules to prevent the occurrence of thrombotic complications. In addition, monitoring this indicator may be useful in determining the effectiveness of preventive therapy. Presentation: 6/3/2024

Fetal MRI findings, etiology, and outcome in prenatally diagnosed schizencephaly.

Schizencephaly is a rare brain anomaly which is increasingly detected in utero. There are limited data on the etiology and outcomes in fetal schizencephaly to guide workup and counselling. We aim to determine the associated imaging findings, etiology, and outcomes in schizencephaly detected in utero. This retrospective cohort study included 22 fetuses with a total of 34 schizencephaly defects identified by keyword search of fetal MRI reports from 1996-2022 followed by image review. Follow-up fetal and postnatal imaging, when available, were reviewed. Data on demographics, etiology, and outcomes were extracted from the electronic medical record. The schizencephaly defect was open in 28/34, most common in the MCA territory (23/34), and commonly involved the frontal (16/34) lobe. Additional intracranial abnormalities were seen in all fetuses including other cortical malformations (CM, 13/22), abnormal posterior fossa (12/22), abnormal corpus callosum (10/20), and intraparenchymal hemorrhage (9/22).The cause of schizencephaly was classified as secondary (as evidenced by intraparenchymal hemorrhage at schizencephaly, monochorionic twin gestation, infection, or maternal/placental risk factor) in 64% (14/22), potentially genetic in 9% (2/22), and unknown in 27% (6/22). Among those liveborn (n=8), the following outcomes were observed: postnatal death (1/8), tube feeding (1/7), shunted hydrocephalus (1/7), epilepsy (4/7). Among those >1 year of age, cerebral palsy (4/5) and speech delay or intellectual disability (3/5) were common. CM remote from schizencephaly was associated with epilepsy (p=0.03). On postnatal imaging, open defects often involuted (8/11) and there were high rates of new/additional findings (4/6). In this cohort, fetal schizencephaly was always associated with additional intracranial abnormalities. In most cases, there was evidence that schizencephaly was likely secondary to prior injury. Imaging characteristics may provide clues regarding neurodevelopmental outcome. Postnatal imaging is crucial in assessing evolution as well as for detection of additional abnormalities. ICH = intracranial hemorrhage; CM = cortical malformation; VM = ventriculomegaly; DGN = deep grey nuclei; SP = septum pellucidum; IPH = intraparenchymal hemorrhage; CC = corpus callosum; PMG = polymicrogyria; PVNH = periventricular nodular heterotopia; TTTS = twin-twin transfusion syndrome; GA = gestational age; CP = cerebral palsy.

Research on abnormal object detection network of computer room inspection robot based on depth vision

Abstract This paper investigates a deep learning-based anomaly object detection network for identifying and alerting on abnormal items within computer room. First, the framework of the data center inspection robot system is outlined, and the anomaly detection task is decomposed. Next, a dataset of abnormal objects based on data center environmental information is established, and augmentation operations are performed on the created dataset. Subsequently, a SqueezeNet network model based on Residual Squeeze Excitation and Atrous Spatial Pyramid Pooling (RSE-ASPP) is proposed to optimize and improve the SqueezeNet network model. Finally, this paper employs transfer learning to address the issue of insufficient data volume. By pre-training on a large-scale dataset and fine-tuning on the constructed dataset, the accuracy and stability of abnormal object recognition can be significantly enhanced. Ultimately, the proposed RSE-ASPP-SqueezeNet network achieves high-precision detection of abnormal items in the data center inspection robot's anomaly detection task.

Pose-oriented scene-adaptive matching for abnormal event detection

For intelligent surveillance systems, abnormal event detection automatically analyses surveillance video sequences and detects abnormal objects or unusual human actions at the frame level. Due to the lack of labelled data, most approaches are semi-supervised based on reconstruction or prediction methods. However, these methods may not generalize well to unseen scene contexts. To address this issue, we present a novel self and mutual scene-adaptive matching method for abnormal event detection. In the framework, we propose synergistic pose estimation and object detection, which effectively integrates human pose and object detection information to improve pose estimation accuracy. Then, the poses are resized to reduce the spatial distance between the source and target domains. The improved pose sequences are further fed into a spatio-temporal graph convolutional network to extract the geometric features. Finally, the features are embedded in a clustering layer to classify action types and compute normality scores. The training data is taken from the training part of common video anomaly detection datasets: UCSD PED1 & PED2, CHUK Avenue, and ShanghaiTech Campus. The proposed framework is evaluated on video sequences with unseen scene contexts in the UCSD PED2 and ShanghaiTech Campus datasets. The detection accuracy and efficiency are also evaluated in detail, and the proposed method for abnormal event detection achieves the highest AUC performance, 84.6%, on the ShanghaiTech Campus dataset and relatively high AUC performance, 96.9% and 74.8%, on UCSD PED2 & PED1 datasets. Compared with other state-of-the-art works, the performance analysis and results confirm the robustness and effectiveness of our proposed framework for cross-scene abnormal event detection.

A-052 Clinical usefulness of ADAMS A1c HA-8190V automatic switching mode for the HbA1c measurement in Hb variants

Abstract Background High-performance liquid chromatography (HPLC) is widely used as a HbA1c routine assay. However, quantification of glycated hemoglobin (HbA1c) is a challenge in patients with hemoglobin (Hb) variants. In this study, we evaluated the clinical utility and analytical performance of new developed ARKRAY ADMAS HA-8190V assay using autoswitch mode, that automatically switches between variant and fast mode depending on patient Hb characteristics, in patients with Hb variant suspected in Korea. Methods The measuring time for HbA1c assay was compared between the HA-8180V fast mode (existing model) and the autoswitch mode of HA-8190V (new model) in 60 normal Hb samples. And the HbA1c measured values was compared bewteen the HA-8180V fast mode and the HA-8190V autoswitch mode in 30 low Hb samples. In addition, flag detection for the abnormal Hb and measured values for 21 Hb variant samples were analyzed using the HA-8180V fast mode and the HA-8190V autoswitch mode. Results It took 50.8 minutes and 39.1 minute to measure 60 normal Hb samples by HA-8180V fast mode and HA-8190V autoswitching mode, respectively. The values could not be measured at the HA-8180V fast mode for 7 of 30 low Hb samples, whereas all samples could be measured at HA-8190V autoswitching mode. For 21 Hb variant samples, HA-8180V fast mode showed abnormal flag detections for only 14 samples, but HA-8190V autoswitch mode reported abnormal variant flags for 20 samples. In the result accuracy evaluation (% difference) based on capillary electrophoresis HbA1c assay (Sebia, Lisses, France) in the 21 Hb variant samples, the % difference of HA-8190 V autoswitch mode was 3.9%, which was superior to HA-8180V fast mode of -11.65%. Conclusions Autoswitch mode of The HA-8190V (automatically switch between variant and fast mode based on patient Hb characteristics) showed more accurate flag detection and HbA1c measurements in patients with Hb variants, along with shorter test time and manual reduction.

Abnormal detection in nuclear security videos based on label-specific autoencoders and reconstruction errors comparison

Abnormal detection in surveillance cameras never fails to be a crucial and challenging task across many fields, aimed at protecting the property safety of citizens and society from abnormal events. Currently, the most common solution for abnormal detection proves to be the combination of autoencoder models and unsupervised training scheme. Due to the scarcity of abnormal samples, all training samples are typically treated as normal samples when constructing the autoencoder calculation models, which generate reconstructed video frames with the same dimension as input data. In this approach, the identification of testing samples is determined by calculating the reconstruction errors between raw and reconstructed video frames. However, such calculation frameworks turn out to be unpractical for nuclear security, as the threshold value of reconstruction errors is overly sensitive, often leading to extreme and unacceptable results. In this manuscript, a novel framework for abnormal detection in nuclear security is proposed. Unlike common unsupervised training scheme, this approach constructs two label-specific autoencoders, while one is trained exclusively with normal samples and the other one is only fed with abnormal samples. Therefore, there is no need to predefine the threshold value for reconstruction errors, as it is replaced by comparing the two label-specific reconstruction errors. Besides, two training schemes for the label-specific autoencoders are proposed and analyzed. Based on a self-collected dataset for nuclear security, the results show that the proposed framework is both practical and appliable, with both training schemes achieving an acceptable accuracy of 0.8182.

Development, Validity, and Reliability of the Primary Eye Care Questionnaire (PECQ) Among Optometrists in Malaysia.

Visual impairment and vision loss are preventable or treatable if there is access to eyecare services for early detection and if proper interventions are given.Optometristsplay a vital role in the early detection of eye and visual-related issues.As healthcare professionals, optometrists are trained to provide comprehensive eye and vision care encompassing dispensing and refraction, detection of abnormalities in the vision and eye, and the rehabilitation of conditions of the visual system, as suggested by the World Council of Optometry.In the current practice of primary eye care, optometrists are not fully utilized and are mostly restricted to refractive correction. Therefore, the optometrists' practice level, scope, and challenges to practice primary eye care must be identified before strengthening optometrists' role in the community. The present study aimed to develop and validate a questionnaireto determine the contributing factors and challenges to the practice of primary eye care by optometrists in Malaysia. The Primary Eye Care Questionnaire (PECQ) was developed through an extensive literature search to identify the domains and to construct the items. The items were then scrutinized further through focus group discussions with experts (n = 10) from various sectors of eye care. Content validation was conducted with ophthalmologists and optometrists as panels (n = 10), followed by face validation (n = 10) by optometrists from the public and private sectors. Construct validation was performed with exploratory factor analysis and reliability was evaluated by using Cronbach's alpha coefficient and the test-retest reliability. In total, the PECQ questionnaire was constructed with 123 itemsunder three main domains: contributing factors, challenges, and continuous education. The content validity index (CVI) for each item (I-CVI) score ranged from 0.9 to 1.0 for each item and the CVI for average scale (S-CVI/Ave) was above 0.8 for each domain, with internal consistency of 0.948 and Cronbach's alpha coefficient of 0.971. The PECQ is a valid and reliable tool for assessing the optometrists' scope of practice, challenges, and contributing factors to the practice of primary eye care.

Mtoto Wetu Growth and Temperature Monitoring System for Tanzania Child Healthcare Clinics

In most government maternal and child health (MCH) clinics in sub-Saharan countries caregivers rely on manual methods for data acquisition, storage, and analysis, leading to challenges such as errors, inefficiencies, and limited accessibility. Recognising the transformative potential of digital technology, this design aims to innovate the child growth monitoring paradigm by introducing a comprehensive Digitized Child Growth Monitoring System. The "Mtoto Wetu Growth and Temperature Monitoring System" is a comprehensive framework designed to monitor and assess the growth and temperature patterns of infants and young children on a monthly basis. This system integrates various components, including anthropometric measurements, temperature monitoring, immunisation schedule, automated growth chart generation, and parental, caregiver, and health care professionals’ involvement, to provide a holistic approach to child health monitoring. Implemented with standardised protocols, training sessions, and robust data management systems, the monitoring system aims to facilitate early detection of growth abnormalities, and health concerns. By promoting early intervention, preventive care measures, parental empowerment, and continuity of care, the Mtoto Wetu system strives to optimise health outcomes, foster parental engagement, and promote health equity among children. The system was able to provide results with a small amount of error compared to the standard weighing machines, tap measure and digital thermometer

Image quilting heuristic compressed sensing video privacy protection coding for abnormal behavior detection in private scenes

For video intelligence applications in private scenes such as home environments, traditional image processing methods are usually based on clear raw data and are prone to privacy leakage. Therefore, our team proposed multilayer compressed sensing (MCS) encoding to reduce image quality for visual privacy protection (VPP). However, the way in which MCS coding is implemented leads to unavoidable information loss. On this basis, inspired by the image quilting (IQ) algorithm, an image quilting heuristic MCS (IQ-MCS) coding method is proposed in this paper to improve the problem of faster information loss in the MCS coding process, which means that a similar privacy protection effect is achieved at lower coding layers, thus obtaining better application performance. To evaluate the level of VPP, a VPP evaluation algorithm is proposed that is more in line with subjective assessment. Finally, a correlation model between the VPP level and the performance of smart applications is established to balance the relationships between them, taking the detection of abnormal human behavior in private scenes as an example. The model can also provide a reference for the evaluation of other privacy protection methods.

USEFULNESS OF VIBRATION ANALYSIS TECHNIQUES AND SENSORS TO IMPROVE THE MONITORING OF INDUSTRIAL EQUIPMENT

The food industry is constantly seeking advanced technologies to improve the efficiency and reliability of machinery, which is critical to improving product quality and manufacturing efficiency. This paper explores the applications of integrating vibration analysis techniques and sensors to improve machine monitoring in the food industry. Vibration analysis is a well-established technique for mechanical fault detection, providing valuable insight into the condition of rotating equipment. By incorporating advanced sensors that can capture a wide range of vibration signals the aim of this research is to improve the accuracy and sensitivity of device health monitoring. The review reviews existing literature on vibration analysis, sensor technology, and applications in the food industry. It explores how traditional vibration analysis methods can be incorporated into advanced sensor technologies such as accelerometers and other sensors. The integration of these technologies will provide a comprehensive understanding of machine behavior, leading to faster abnormal detection and predictive maintenance.

Chromosomal Abnormalities Detected by Chromosomal Microarray Analysis and Karyotype in Fetuses with Ultrasound Abnormalities.

Chromosomal microarray analysis (CMA) is a first-line test to assess the genetic etiology of fetal ultrasound abnormalities. The aim of this study was to evaluate the effectiveness of CMA in detecting chromosomal abnormalities in fetuses with ultrasound abnormalities, including structural abnormalities and non-structural abnormalities. A retrospective study was conducted on 368 fetuses with abnormal ultrasound who received interventional prenatal diagnosis at Meizhou People's Hospital from October 2022 to December 2023. Samples of villi, amniotic fluid, and umbilical cord blood were collected according to different gestational weeks, and karyotype and CMA analyses were performed. The detection rate of chromosomal abnormalities in different ultrasonic abnormalities was analyzed. There were 368 fetuses with abnormal ultrasound, including 114 (31.0%) with structural abnormalities, 225 (61.1%) with non-structural abnormalities, and 29 (7.9%) with structural combined with non-structural abnormalities. The detection rate of aneuploidy and pathogenic (P)/likely pathogenic (LP) copy number variations (CNVs) of CMA in fetuses with structural abnormalities was 5.26% (6/114), the detection rate of karyotype was 2.63% (3/114), and the additional diagnosis rate of CMA was 2.63%. In the fetuses with ultrasonic non-structural abnormalities, the detection rate of karyotype was 6.22% (14/225), the detection rate of aneuploidy and P/LP CNVs in fetuses with ultrasonic structural abnormalities was 9.33% (21/225), and the additional diagnosis rate of CMA was 3.11%. There was no significant difference in chromosome abnormality detection rate of CMA among structural abnormality, non-structural abnormality, and structural abnormality combined with non-structural abnormality groups (5.3%, 9.3%, and 13.8%, p = 0.241), also among multiple ultrasonic abnormality and single ultrasonic abnormality groups (14.8%, and 7.3%, p = 0.105). CMA can significantly improve the detection rate of genetic abnormalities in prenatal diagnosis of ultrasonic abnormal fetuses compared with karyotype analysis. CMA is a more effective tool than karyotyping alone in detecting chromosomal abnormalities in fetuses with ultrasound abnormalities.

Heart Sound Abnormality Detection From Multi-Institutional Collaboration: Introducing a Federated Learning Framework.

Early diagnosis of cardiovascular diseases is a crucial task in medical practice. With the application of computer audition in the healthcare field, artificial intelligence (AI) has been applied to clinical non-invasive intelligent auscultation of heart sounds to provide rapid and effective pre-screening. However, AI models generally require large amounts of data which may cause privacy issues. Unfortunately, it is difficult to collect large amounts of healthcare data from a single centre. In this study, we propose federated learning (FL) optimisation strategies for the practical application in multi-centre institutional heart sound databases. The horizontal FL is mainly employed to tackle the privacy problem by aligning the feature spaces of FL participating institutions without information leakage. In addition, techniques based on deep learning have poor interpretability due to their "black-box" property, which limits the feasibility of AI in real medical data. To this end, vertical FL is utilised to address the issues of model interpretability and data scarcity. Experimental results demonstrate that, the proposed FL framework can achieve good performance for heart sound abnormality detection by taking the personal privacy protection into account. Moreover, using the federated feature space is beneficial to balance the interpretability of the vertical FL and the privacy of the data. This work realises the potential of FL from research to clinical practice, and is expected to have extensive application in the federated smart medical system.

Exploring Techniques for Abnormal Event Detection in Video Surveillance

This study explores and evaluates the effectiveness of various abnormal event detection techniques in video surveillance, addressing challenges such as intrusions, accidents, and suspicious activities. Through a systematic review of related papers, the study reveals the prevalence of traditional methods like background subtraction and motion detection despite their limitations in complex scenarios. It highlights the increasing use of deep learning techniques, particularly CNNs and RNNs, which show promise but require substantial labeled data. The findings underscore the importance of selecting proper detection techniques based on specific surveillance scenarios and emphasize the need for extensive labeled datasets for deep learning methods. The originality of this study lies in its comprehensive review and comparison of various abnormal event detection techniques, providing valuable insights and practical implications for advancing video surveillance systems.

Selective segmentation of brain abnormalities in colour MRI images using variational model

Early detection of brain abnormalities is vital for enhancing patient outcomes and survival rates. However, accurately identifying and segmenting these abnormalities from MRI images remains a persistent challenge. This study assesses the efficacy of the Selective Local Image Fitting (SLIF) model in segmenting brain abnormalities from colour MRI images and compares its performance with converted greyscale counterparts. The rationale behind this comparison stems from standard practice in image segmentation, where colour images are often converted to greyscale before the segmentation task. Converting the image might degrade data by diminishing its dimensions, potentially affecting segmentation computations. This study intends to evaluate the influence of colour information on segmentation accuracy and efficiency by directly assessing the SLIF model on both colour and converted greyscale images. Segmentation accuracy was evaluated using metrics such as the Dice Similarity Coefficient (DSC), Matthews Correlation Coefficient (MCC), and Intersection-over-Union (IoU). Efficiency was determined by measuring the average elapsed processing time. Experimental results demonstrate that colour MRI brain images outperform their converted greyscale counterparts in segmentation accuracy, as colour providing essential supplementary information for precise abnormality delineation. Despite a slight increase in average elapsed processing time for colour images, the enhanced accuracy justifies this trade-off. These findings emphasize the importance of colour MRI in enhancing diagnostic accuracy, especially in detecting brain abnormalities. This study can be extended in future work to evaluate the segmentation accuracy and efficiency of brain abnormalities in 3D colour and greyscale MRI images.

Enhancing pap smear image classification: integrating transfer learning and attention mechanisms for improved detection of cervical abnormalities

Cervical cancer remains a major global health challenge, accounting for significant morbidity and mortality among women. Early detection through screening, such as Pap smear tests, is crucial for effective treatment and improved patient outcomes. However, traditional manual analysis of Pap smear images is labor-intensive, subject to human error, and requires extensive expertise. To address these challenges, automated approaches using deep learning techniques have been increasingly explored, offering the potential for enhanced diagnostic accuracy and efficiency. This research focuses on improving cervical cancer detection from Pap smear images using advanced deep-learning techniques. Specifically, we aim to enhance classification performance by leveraging Transfer Learning (TL) combined with an attention mechanism, supplemented by effective preprocessing techniques. Our preprocessing pipeline includes image normalization, resizing, and the application of Histogram of Oriented Gradients (HOG), all of which contribute to better feature extraction and improved model performance. The dataset used in this study is the Mendeley Liquid-Based Cytology (LBC) dataset, which provides a comprehensive collection of cervical cytology images annotated by expert cytopathologists. Initial experiments with the ResNet model on raw data yielded an accuracy of 63.95%. However, by applying our preprocessing techniques and integrating an attention mechanism, the accuracy of the ResNet model increased dramatically to 96.74%. Further, the Xception model, known for its superior feature extraction capabilities, achieved the best performance with an accuracy of 98.95%, along with high precision (0.97), recall (0.99), and F1-Score (0.98) on preprocessed data with an attention mechanism. These results underscore the effectiveness of combining preprocessing techniques, TL, and attention mechanisms to significantly enhance the performance of automated cervical cancer detection systems. Our findings demonstrate the potential of these advanced techniques to provide reliable, accurate, and efficient diagnostic tools, which could greatly benefit clinical practice and improve patient outcomes in cervical cancer screening.

Concept-based AI interpretability in physiological time-series data: Example of abnormality detection in electroencephalography

Background and ObjectiveDespite recent performance advancements, deep learning models are not yet adopted in clinical practice on a wide scale. The intrinsic intransparency of such systems is commonly cited as one major reason for this reluctance. This has motivated methods that aim to provide explanations of model functioning. Known limitations of feature-based explanations have led to an increased interest in concept-based interpretability. Testing with Concept Activation Vectors (TCAV) employs human-understandable, abstract concepts to explain model behavior. The method has previously been applied to the medical domain in the context of electronic health records, retinal fundus images and magnetic resonance imaging. MethodsWe explore the usage of TCAV for building interpretable models on physiological time series, using an example of abnormality detection in electroencephalography (EEG). For this purpose, we adopt the XceptionTime model, which is suitable for multi-channel physiological data of variable sizes. The model provides state-of-the-art performance on raw EEG data and is publically available. We propose and test several ideas regarding concept definition through metadata mining, using additional labeled EEG data and extracting interpretable signal characteristics in the form of frequencies. By including our own hospital data with analog labeling, we further evaluate the robustness of our approach. ResultsThe tested concepts show a TCAV score distribution that is in line with the clinical expectations, i.e. concepts known to have strong links with EEG pathologies (such as epileptiform discharges) received higher scores than the neutral concepts (e.g. sex). The scores were consistent across the applied concept generation strategies. ConclusionsTCAV has the potential to improve interpretability of deep learning applied to multi-channel signals as well as to detect possible biases in the data. Still, further work on developing the strategies for concept definition and validation on clinical physiological time series is needed to better understand how to extract clinically relevant information from the concept sensitivity scores.

Cystic Fibrosis: A Review of Current Knowledge, Treatment Methods and Diagnostics

Introduction: Cystic fibrosis (CF) is an autosomal recessive genetic disease caused by a mutation in the CFTR gene. Characteristic symptoms include persistent lung infection, pancreatic insufficiency, and elevated sweat chloride, although many patients have mild or atypical symptoms. Diagnosis of CF is based primarily on the detection of genetic and/or functional abnormalities in the CFTR gene. Treatment includes pancreatic enzyme replacement therapy, mucolytic drugs, respiratory physiotherapy, and antibiotic therapy. Purpose of the work: This study aims to review and characterize the clinical and pathophysiological aspects of cystic fibrosis. Materials and methods: A comprehensive analysis of research papers available on PubMed, Google Scholar, Web of Science, Embase and Scopus was undertaken using the searchterms encompassing the following keywords: Cystic fibrosis, CFTR, Chloride Channels, Lung Inflammation, Diagnosis, Gene therapy, Treatment. Results: Cystic fibrosis is the most common, multisystem, life-threatening recessive disease in the white race. Although many organs are affected, the leading cause of morbidity and mortality is lung disease. Treatment requires pharmacotherapy, extensive physiotherapy, and nutritional support. In recent years, there has been significant progress in the diagnosis and treatment of cystic fibrosis, and the average survival has increased from a few to almost 50 years.

Echocardiography In Acute Stroke - A Precious Resource

Abstract Background Echocardiography (Echo) is frequently utilized as part of the work up for ischemic stroke. National guidelines suggest using echo to assess for source of unexplained stroke “if detection of a structural cardiac abnormality would prompt a change of management”. This study aims to describe the use of Echocardiography post-stroke in Mayo University Hospital (MUH). Methods The study population was derived from a list of strokes admitted to MUH from July to December 2023 which was compiled by the Stroke ANP. The radiology system was consulted to assess time to echo and findings. Results 86 ischemic strokes were identified. 77 (89%) had echocardiography ordered as inpatient. 61 of 77 echoes ordered were filmed during admission. The mean wait for echo was 6.4 days. No PFO or LV thrombus was detected. In 52% (n=32) of echo’s performed for work up of aetiology of stroke there was no mention of intra-atrial septum (IAS) in the report. 24% had Modified Rankin Scale on discharge of 4 or above. Of these, 75% had echo ordered as inpatient despite 60% already having atrial fibrillation diagnosed. Conclusion Echocardiography appeared to be an over-utilised resource in ischemic strokes in MUH. The vast majority of patients with stroke, even if frail, elderly, or severely disabled, had echocardiography ordered despite national guidelines suggesting usage of echo only when diagnosis of structural heart disease was likely to change management. The intra-atrial septum wasn’t even mentioned in the majority of studies. The wait for echocardiography appears to be extending bed days for patients putting strain on hospital inpatient capacity and emergency departments. A more nuanced approach to ordering echo in the setting of acute stroke could be advocated for, this will take a collaborative effort between Cardiology and Stroke Medicine in order to prioritise that will most benefit from this precious resource.

Automated grading of diabetic retinopathy and Radiomics analysis on ultra-wide optical coherence tomography angiography scans

Diabetic retinopathy (DR), a progressive condition due to diabetes that can lead to blindness, is typically characterized by a number of stages, including non-proliferative (mild, moderate and severe) and proliferative DR. These stages are marked by various vascular abnormalities, such as intraretinal microvascular abnormalities (IRMA), neovascularization (NV), and non-perfusion areas (NPA). Automated detection of these abnormalities and grading the severity of DR are crucial for computer-aided diagnosis. Ultra-wide optical coherence tomography angiography (UW-OCTA) images, a type of retinal imaging, are particularly well-suited for analyzing vascular abnormalities due to their prominence on these images. However, accurate detection of abnormalities and subsequent grading of DR is quite challenging due to noisy data, presence of artifacts, poor contrast and subtle nature of abnormalities. In this work, we aim to develop an automated method for accurate grading of DR severity on UW-OCTA images. Our method consists of various components such as UW-OCTA scan quality assessment, segmentation of vascular abnormalities and grading the scans for DR severity. Applied on publicly available data from Diabetic retinopathy analysis challenge (DRAC 2022), our method shows promising results with a Dice overlap metric and recall values of 0.88 for abnormality segmentation, and the coefficient-of-agreement (κ) value of 0.873 for DR grading. We also performed a radiomics analysis, and observed that the radiomics features are significantly different for increasing levels of DR severity. This suggests that radiomics could be used for multimodal grading and further analysis of DR, indicating its potential scope in this area.

Bearing RUL prediction and fault diagnosis system based on parallel multi-scale MIMT lightweight model

Abstract In actual industrial production, the importance of safety production is increasingly prominent, and the degradation and failure of machinery and equipment are potential sources of safety hazards. Therefore, there is a growing trend towards real-time monitoring, prediction, and diagnosis of industrial equipment to prevent unpredictable impacts on life and property safety caused by sudden failures. To address this issue, this paper proposes a real-time degradation anomaly detection based on parallel multiscale autoencoders and a lightweight model of parallel multiscale multi-input multi-task for bearing Remaining Useful Life (RUL) prediction and fault diagnosis systems. Firstly, the multiscale autoencoder method is used to simulate actual working conditions and reconstruct the original vibration signals to build abnormal degradation detection intervals. The [0, $\mu$ +3$\sigma$] interval is utilized to judge abnormal degradation based on reconstruction errors, and the First Predict Timepoint (FPT) is determined adaptively. Secondly, a method for constructing dimensionless auxiliary datasets is proposed, which adopts a multi-input form based on deep separable convolution for feature extraction of original vibration signals, kurtosis, and peak values to improve the prediction and diagnosis performance of the lightweight model. Finally, a multi-task output method combining clustering and regression is employed to achieve RUL prediction and fault diagnosis of bearings. The proposed method overcomes the problems existing in traditional bearing RUL prediction and diagnosis methods and possesses theoretical innovation and engineering practicality. Validation on two bearing datasets confirms the effectiveness of the proposed method.