Imaging Abnormalities Research Articles

Specific brain MRI features of constitutional mismatch repair deficiency syndrome in children with high-grade gliomas.

Children with constitutional mismatch repair deficiency (CMMRD) syndrome have an increased risk of high-grade gliomas (HGG), and brain imaging abnormalities. This study analyzes brain imaging features in CMMRD syndrome children versus those with HGG without CMMRD. Retrospective comparative analysis of brain imaging in 30 CMMRD children (20 boys, median age eight years, 22 with HGG), seven with Lynch syndrome (7 HGG), 39 with type 1 neurofibromatosis (NF1) (four with HGG) and 50 with HGG without MMR or NF1 pathogenic variant ("no-predisposition" patients). HGG in CMMRD and Lynch patients were predominantly hemispheric (versus midline) compared to NF1 and no-predisposition patients (91% and 86%, vs 25% and 54%, p = 0.004). CMMRD-associated tumors often had ill-defined boundaries (p = 0.008). All CMMRD patients exhibited at least one developmental venous anomaly (DVA), versus 14%, 10%, and 6% of Lynch, NF1, and no-predisposition patients (p < 0.0001). Multiple DVAs were observed in 83% of CMMRD patients, one NF1 patient (3%), and never in other groups (p < 0.0001). Cavernomas were discovered in 21% of CMMRD patients, never in other groups (p = 0.01). NF1-like focal areas of high T2-FLAIR signal intensity (FASI) were more prevalent in CMMRD patients than in Lynch or no-predisposition patients (50%, vs 20% and 0%, respectively, p < 0.0001). Subcortical and ill-limited FASI, possibly involving the cortex, were specific to CMMRD (p < 0.0001) and did not evolve in 93% of patients (13/14). Diffuse hemispherically located HGG associated with multiple DVAs, cavernomas, and NF1-like or subcortical FASI strongly suggests CMMRD syndrome compared to children with HGG in other contexts. The radiologic suggestion of CMMRD syndrome when confronted with HGGs in children may prompt genetic testing. This can influence therapeutic plans. Therefore, imaging features could potentially be incorporated into CMMRD testing recommendations. Using imaging to detect CMMRD syndrome early may improve patient care. CMMRD features include: hemispheric HGG with multiple developmental venous anomalies and NF1-like or subcortical areas with high T2-FLAIR intensity. We propose novel imaging features to improve the identification of potential CMMRD patients.

MR Imaging Findings in Anti-Leucine-Rich Glioma Inactivated Protein 1 Encephalitis: A Systematic Review and Meta-analysis.

Antibodies against leucine-rich glioma inactivated protein 1 (LGI1) constitute a common form of autoimmune encephalitis. On MR imaging, it may show T2 FLAIR hyperintensities of the medial temporal lobe (T2 FLAIR-MTL), involve the basal ganglia, or be unremarkable. We performed a systematic review and meta-analysis to obtain prevalence estimates of abnormal findings on MR imaging in anti-LGI1 encephalitis. A human brain map of the LGI1 microarray gene expression was derived from the Allen Human Brain Atlas. PubMed and Web of Science were searched with the terms "LGI1" and "encephalitis" from inception to April 7, 2022. Thirty-one research publications, encompassing case series and retrospective cohort and case-control studies, with >10 patients with anti-LGI1 encephalitis and MR imaging data were included. Pooled prevalence estimates were calculated using Freeman-Tukey double-arcsine transformation. Meta-analysis used DerSimonian and Laird random effects models. Of 1318 patients in 30 studies, T2 FLAIR-MTL hyperintensities were present in 54% (95% CI, 0.48-0.60; I2 = 76%). Of 394 patients in 13 studies, 27% showed bilateral (95% CI, 0.19-0.36; I2 = 71%) and 24% unilateral T2 FLAIR-MTL abnormalities (95% CI, 0.17-0.32; I2 = 61%). Of 612 patients in 15 studies, basal ganglia abnormalities were present in 10% (95% CI, 0.06-0.15; I2 = 67%). LGI1 expression was highest in the amygdala, hippocampus, and caudate nucleus. Only part of the spectrum of MR imaging abnormalities in anti-LGI1 encephalitis could be included in a meta-analysis. MR imaging findings were not the main outcomes in most studies, limiting available information. I2 values ranged from 62% to 76%, representing moderate-to-large heterogeneity. T2 FLAIR-MTL hyperintensities were present in around one-half of patients with anti-LGI1. The prevalence of unilateral and bilateral presentations was similar, suggesting unilaterality should raise the suspicion of this disease in the appropriate clinical context. Around 10% of patients showed basal ganglia abnormalities, indicating that special attention should be given to this region. LGI1 regional expression coincided with the most frequently reported abnormal findings on MR imaging. Regional specificity might be partially determined by expression levels of the target protein.

The use of artificial intelligence in radiology: new possibilities for diagnostic imaging. A literature review

Introduction and purpose: Rapid advances in technology enable innovative solutions to be implemented in modern medicine, relieving healthcare workers by speeding up diagnosis and improving the quality of treatment. The subject of this review is Artificial Intelligence (AI), an innovative form of help in the daily practice of doctors. It offers the opportunity to relieve healthcare workers by accelerating the diagnosis process and improving the quality and effectiveness of treatment. The aim of this paper is to present the current state of knowledge, assess the effectiveness of algorithms in the recognition and interpretation of abnormalities in medical images compared to specialists in radiology and discuss the challenges associated with the implementation of AI in various medical specialties. Brief description of the state of knowledge: Artificial intelligence, especially through machine learning and deep learning techniques, has found wide application in radiology. Many facilities around the world use advanced AI systems in the day-to-day work of radiologists, including the Mayo Clinic, Massachusetts General Hospital, and the University of Tokyo Hospital. Summary: Studies show that while AI algorithms can perform worse than radiologists in some areas, they are at the forefront of others, especially in detecting subtle abnormalities. The effective implementation of artificial intelligence requires addressing regulatory, ethical, and training issues. Despite these challenges, artificial intelligence has the potential to play a key role in the future of radiology and revolutionize medical practice, opening new perspectives and improving the quality of health care.

Unveiling Opportunities for Intervention: A Prospective Cohort Study Investigating the Clinical Significance of Diffusion-Weighted Imaging (DWI)-Fluid-Attenuated Inversion Recovery (FLAIR) Mismatch Beyond the Window Period in Acute Ischemic Stroke.

Acute ischemic stroke causes irreversible damage to the brain parenchymasurrounded by salvageable tissue known as the ischemic penumbra.Magnetic resonance imaging (MRI), particularly the mismatch between abnormal diffusion-weighted imaging (DWI) signals and normal fluid-attenuated inversion recovery (FLAIR) signals, plays a critical role in detecting ischemic penumbra. It also allows for the identification of patients who may benefit from reperfusion therapy. Hence, this prospective cohort study aimed to explore the correlation between DWI-FLAIR mismatch and clinical outcomes in acute ischemic stroke patients, specifically those with delayed or uncertain symptom onset, offering potential insights into reperfusion therapy. A total of 38 thrombotic stroke patients aged above 18 were included in this prospective cohort study. Baseline data, including demographics, lifestyle factors, and medical history, were recorded. DWI-FLAIR mismatch was evaluated through brain MRI within 4.5 hours to 12 hours of symptom onset. Of the cohort, 63.2% were males, predominantly in the 61-70 age group. Smoking and alcohol consumption were reported by 15.79% each. DWI-FLAIR mismatch was present in 20 out of 38 subjects. No statistically significant differences were noted in the mean National Institutes of Health Stroke Scale (NIHSS) and Modified Rankin Scale (MRS) scores between subjects with and without DWI-FLAIR mismatch. Thrombolysis in wake-up stroke subjects demonstrated a substantial reduction in mean MRS at discharge (1.29±0.95) and at six to eight weeks (1.71±1.11), suggesting potential benefits on functional outcomes. The prevalence of DWI-FLAIR mismatch was seen in the majority of patients beyond their window period and also showed beneficiary outcomes with a mean reduction in NHISS and MRS scores following thrombolysis.

Cerebral amyloid-β-related angiitis and iatrogenic cerebral amyloid angiopathy-related vasculitis: Implications for amyloid-related imaging abnormalities.

Cerebral amyloid-β-related angiitis and iatrogenic cerebral amyloid angiopathy-related vasculitis: Implications for amyloid-related imaging abnormalities.

Medical Image Classification using a Many to Many Relation, Multilayered Fuzzy Systems and AI

One of the research gaps in the medical sciences is the study of orphan diseases or rare diseases, due to limited data availability of rare diseases. Our previous study addressed this successfully by developing an Artificial Intelligence (AI)-based medical image classification method using a multilayer fuzzy approach (MFA), for detecting and classifying image abnormalities for large and very small datasets. A fuzzy system is an AI system used to handle imprecise data. There are more than three types of fuzziness in any image data set: 1) due to a projection of a 3D object on a 2D surface, 2) due to the digitalization of the scan, and 3) conversion of the digital image to grayscale, and more. Thus, this was referred to in the previous study as a multilayer fuzzy system, since fuzziness arises from multiple sources. The method used in MFA involves comparing normal images containing abnormalities with the same kind of image without abnormalities, yielding a similarity measure percentage that, when subtracted from a hundred, reveals the abnormality. However, relying on a single standard image in the MFA reduces efficiency, since images vary in contrast, lighting, and patient demographics, impacting similarity percentages. To mitigate this, the current study focused on developing a more robust medical image classification method than MFA, using a many-to-many relation and a multilayer fuzzy approach (MCM) that employs multiple diverse standard images to compare with the abnormal image. For each abnormal image, the average similarity was calculated across multiple normal images, addressing issues encountered with MFA, and enhancing versatility. In this study, an AI-based method of image analysis automation that utilizes fuzzy systems was applied to a cancer data set for the first time. MCM proved to be highly efficient in detecting the abnormality in all types of images and sample sizes and surpassed the gold standard, the convolutional neural network (CNN), in detecting the abnormality in images from a very small data set. Moreover, MCM detects and classifies abnormality without any training, validation, or testing steps for large and small data sets. Hence, MCM may be used to address one of the research gaps in medicine, which detects, quantifies, and classifies images related to rare diseases with small data sets. This has the potential to assist a physician with early detection, diagnosis, monitoring, and treatment planning of several diseases, especially rare diseases.

Clinical characteristics analysis of 4 cases with acute flaccid myelitis in children

Objective: To summarize the clinical manifestations, diagnosis, treatment and prognosis of acute flaccid myelitis (AFM) in children. Methods: Clinical characteristics of 4 AFM cases from Department of Neurology, Children's Hospital Affiliated to Capital Institute of Pediatrics, from September 2018 to November 2022, were analyzed retrospectively. Results: The age of 4 children with AFM was 7 years, 4 years and 3 months, 7 years and 1 month, 6 years and 5 months, respectively. There were 2 boys and 2 girls. Prodromal infection status showed 3 children of respiratory tract infection and 1 child of digestive tract infection. The main manifestation was asymmetrical limb weakness after infection, and the affected limb range was from monoplegia to quadriplegia. Cranial nerve injury was involved in 1 child, no encephalopathy. Magnetic resonance imaging in the spinal cord of all 4 children showed long T1 and T2 signals, mainly involving gray matter. Cerebrospinal fluid cell-protein separation was observed in 2 children. Pathogen detected in 1 child pharyngeal swab was enterovirus D68. Antibody IgM to adenovirus was positive in the blood of 1 child. Antibody IgG against Echo and Coxsackie B virus were positive in the blood of another child. After glucocorticoid, human immunoglobulin or simple symptomatic treatment and at the same time under later rehabilitation training, muscle strength recovered to different degrees, but there were disabilities left in 3 children. Conclusions: AFM should be considered in children with acute and asymmetrical flaccid paralysis accompanied by abnormal magnetic resonance imaging signal in the central region of spinal cord, especially post-infection. The effective treatment is limited and the prognosis is poor.

Generative adversarial network-based reconstruction of healthy anatomy for anomaly detection in brain CT scans.

To help radiologists examine the growing number of computed tomography (CT) scans, automatic anomaly detection is an ongoing focus of medical imaging research. Radiologists must analyze a CT scan by searching for any deviation from normal healthy anatomy. We propose an approach to detecting abnormalities in axial 2D CT slice images of the brain. Although much research has been done on detecting abnormalities in magnetic resonance images of the brain, there is little work on CT scans, where abnormalities are more difficult to detect due to the low image contrast that must be represented by the model used. We use a generative adversarial network (GAN) to learn normal brain anatomy in the first step and compare two approaches to image reconstruction: training an encoder in the second step and using iterative optimization during inference. Then, we analyze the differences from the original scan to detect and localize anomalies in the brain. Our approach can reconstruct healthy anatomy with good image contrast for brain CT scans. We obtain median Dice scores of 0.71 on our hemorrhage test data and 0.43 on our test set with additional tumor images from publicly available data sources. We also compare our models to a state-of-the-art autoencoder and a diffusion model and obtain qualitatively more accurate reconstructions. Without defining anomalies during training, a GAN-based network was used to learn healthy anatomy for brain CT scans. Notably, our approach is not limited to the localization of hemorrhages and tumors and could thus be used to detect structural anatomical changes and other lesions.

Attention-deficit/hyperactivity disorder (ADHD) in adults: a multilayered approach to a serious disorder of inattention to the future.

Attention-deficit/hyperactivity disorder (ADHD) affects people of all ages, yet its presentation varies as the person matures and social demands increase from childhood into adulthood. Interestingly, when analyzing the disorder in adults, it is not immediately clear what the 'attention deficit' in the ADHD denomination exactly means. Adults with ADHD have a broad range of difficulties, far beyond an attentional deficit, that impact negatively their social functioning and often lead to failures in all walks of life. Therefore, in this review, I attempt to reconcile the notion of attention deficit with the protean manifestations of ADHD in adults through a proposal that ADHD symptoms have as a common denominator an inattention to the future. I build this construct through a multilayered approach, progressing from the epidemiological and clinical considerations for Diagnostic and Statistical Manual of Mental Disorders (DSM) diagnosis, to a deeper understanding of the disorder, discussing how these patients fail to anchor the present into the future (i.e., to be attentive to future consequences), thus failing to approximate future goals from present action. Integrating cognitive observations with imaging abnormalities, it is possible to propose that ADHD in adults is perhaps the most prevalent frontal lobe disorder in humans, ultimately impacting upon psychosocial management and treatment strategies.

Blake's pouch cyst-like imaging abnormalities in idiopathic normal pressure hydrocephalus

ObjectiveBlake's pouch cyst is a cause of congenital non-communicating hydrocephalus. Recent study revealed that adult onset communicating normal pressure hydrocephalus sometimes accompanied by Blake's pouch cyst-like imaging abnormalities. However, relationships between two disease entities remains uncovered. The aim of this study was to investigate the prevalence of Blake's pouch cyst-like imaging abnormalities in idiopathic normal pressure hydrocephalus and their impact on the clinical features. MethodsFirst, we investigated the existence of Blake's pouch cyst-like imaging abnormalities in 46 idiopathic normal pressure hydrocephalus patients (mean age 76.8 ± 5.2 years old; 28 women) and 247 normal controls (mean age 65.2 ± 4.0 years old; 120 women). Second, the demographic, clinical and imaging characteristics were compared between idiopathic normal pressure hydrocephalus with and without Blake's pouch cyst-like imaging abnormalities. ResultsBlake's pouch cyst-like imaging abnormalities were more prevalent in idiopathic normal pressure hydrocephalus patients (17.4 %) than in the normal population (2.4 %). Idiopathic normal pressure hydrocephalus with Blake's pouch cyst-like imaging abnormalities showed a trend toward a younger onset than those without Blake's pouch cyst-like imaging abnormalities, while clinical features, including the response to cerebrospinal fluid tap test and shunt treatment, did not differ between two groups. ConclusionsIncomplete cerebrospinal fluid obstruction associated with Blake's pouch cyst-like imaging abnormalities may be a cause of idiopathic normal pressure hydrocephalus.

APOLLOE4 Phase 3 study of oral ALZ-801/valiltramiprosate in APOE ε4/ε4 homozygotes with early Alzheimer's disease: Trial design and baseline characteristics.

The approved amyloid antibodies for early Alzheimer's disease (AD) carry a boxed warning about the risk of amyloid-related imaging abnormalities (ARIAs) that are highest in apolipoprotein E (APOE) ε4/ε4 homozygotes. ALZ-801/valiltramiprosate, an oral brain-penetrant amyloid beta oligomer inhibitor is being evaluated in APOE ε4/ε4 homozygotes with early AD. This Phase 3 randomized, double-blind, placebo-controlled, 78-week study of ALZ-801 administered as 265mg twice per day tablets, enrolled 50- to 80-year-old homozygotes with Mini-Mental State Examination (MMSE) ≥ 22 and Clinical Dementia Rating-Global Score 0.5 or 1.0. The study is powered to detect a 2.0 to 2.5 drug-placebo difference on the Alzheimer's Disease Assessment Scale 13-item Cognitive subscale primary outcome with 150 subjects/arm. The key secondary outcomes are Clinical Dementia Rating-Sum of Boxes and Instrumental Activities of Daily Living; volumetric magnetic resonance imaging and fluid biomarkers are additional outcomes. The APOLLOE4 Phase 3 trial enrolled 325 subjects with a mean age of 69 years, 51% female, MMSE 25.6, and 65% mild cognitive impairment. Topline results are expected in 2024. APOLLOE4 is the first disease-modification AD trial focused on APOE ε4/ε4 homozygotes. Oral ALZ-801 has the potential to be the first effective and safe anti-amyloid treatment for the high-risk APOE ε4/ε4 population. The APOLLOE4 Phase 3, placebo-controlled, 78-week study is designed to evaluate the efficacy and safety of ALZ-801 265mg twice per day in early Alzheimer's disease (AD) subjects with the apolipoprotein E (APOE) ε4/ε4 genotype.The enrolled early AD population (N=325) has 51% females, a mean age = 69 years, and a mean Mini-Mental State Examination = 25.6, with the majority being mild cognitive impairment subjects, a similar disease stage to the lecanemab Phase 3 AD trial (Clarity AD).The primary outcome is the cognitive Alzheimer's Disease Assessment Scale 13-item Cognitive subscale, with two functional measures as key secondary outcomes (Clinical Dementia Rating-Sum of Boxes, Amsterdam-Instrumental Activities of Daily Living), and with hippocampal volume and fluid biomarkers as additional outcomes.The study is unique in allowing a large number of microhemorrhages or siderosis at baseline magnetic resonance imaging, lesions that indicate concomitant cerebral amyloid angiopathy (CAA).At baseline, 32% of the enrolled population had at least 1 microhemorrhage, 24% had 1 to 4, and 8% had > 4 microhemorrhages; 10% had at least 1 siderosis lesion; with more males than females having microhemorrhages (63% vs. 37%) and siderosis (68% vs. 32%).Study results will become available in the second half of 2024 and, if positive, ALZ-801 may become the first oral drug to demonstrate a favorable benefit/risk profile in APOE ε4/ε4 AD subjects.

Multi-Grained Radiology Report Generation With Sentence-Level Image-Language Contrastive Learning.

The automatic generation of accurate radiology reports is of great clinical importance and has drawn growing research interest. However, it is still a challenging task due to the imbalance between normal and abnormal descriptions and the multi-sentence and multi-topic nature of radiology reports. These features result in significant challenges to generating accurate descriptions for medical images, especially the important abnormal findings. Previous methods to tackle these problems rely heavily on extra manual annotations, which are expensive to acquire. We propose a multi-grained report generation framework incorporating sentence-level image-sentence contrastive learning, which does not require any extra labeling but effectively learns knowledge from the image-report pairs. We first introduce contrastive learning as an auxiliary task for image feature learning. Different from previous contrastive methods, we exploit the multi-topic nature of imaging reports and perform fine-grained contrastive learning by extracting sentence topics and contents and contrasting between sentence contents and refined image contents guided by sentence topics. This forces the model to learn distinct abnormal image features for each specific topic. During generation, we use two decoders to first generate coarse sentence topics and then the fine-grained text of each sentence. We directly supervise the intermediate topics using sentence topics learned by our contrastive objective. This strengthens the generation constraint and enables independent fine-tuning of the decoders using reinforcement learning, which further boosts model performance. Experiments on two large-scale datasets MIMIC-CXR and IU-Xray demonstrate that our approach outperforms existing state-of-the-art methods, evaluated by both language generation metrics and clinical accuracy.

Artificial Intelligence in Healthcare: Implications, Challenges, and Future Prospects

Artificial intelligence is a technology that enables machines, especially computer systems, to mimic human intelligence and problem-solving abilities. This technology is being used nowadays in many areas of the healthcare industry, from scheduling online appointments to robot-assisted, safer surgeries with better patient outcomes. Just like a physician, AI can record a patient's history, signs and symptoms, and lab results, leading to an accurate diagnosis and a proper treatment plan, but in no time. This is just a small example of the implications of AI in the sector. Other areas of medicine are also being benefited by this innovation, like gastroentology, radiology, surgery, and preventive medicine. The development of advanced algorithms has reduced the burden on radiologists by helping them identify abnormal images and pick up malignant lesions with minimal diagnostic errors. Similarly, AI-assisted colonoscopy can help identify malignant and benign polyps. AI-powered systems, like Google's DeepMind Health, may identify malignant growth in mammograms and diabetic retinopathy, which can aid in early detection and treatment.

The role for endobronchial biopsy in the era of endobronchial ultrasound guided transbronchial needle aspiration for the diagnosis of sarcoidosis: a single center experience.

Endobronchial biopsy (EBBX) has been reported to increase diagnostic yield for pulmonary sarcoidosis. The purpose of this study is to investigate the diagnostic yield for EBBX following endobronchial ultrasound guided transbronchial needle aspiration (EBUS-TBNA). We identified a cohort of patients in the University of Minnesota Sarcoidosis Registry who had EBBx and EBUS-TBNA as part of workup for abnormal chest imaging. Data regarding demographics, biopsy approach and technique were recorded. Our cohort included 37 patients (53.24±9.5, Male, 22±0.57; 3.8% were African American). In these patients who had EBBX, EBUS-TBNA was performed in 100% of patients and TBBX was performed in 2 patients (5%). EBBX was positive in 9 patients (24%) and EBUS-TBNA was positive in 34 patients (92%). TBBX was diagnostic in one of two patients. EBBX was the only diagnostic tissue in 3 of the 37 patients (8%). Conclusion: The diagnostic yield of EBBX is lower than previously reported, with only 8% of EBBXs demonstrating granulomatous inflammation. However, instrumentation used for obtaining EBBX as well as the presence of visible lesions does influence the diagnostic yield. Studies with adequate power are needed before implementing changes in clinical practice. When performed alongside EBUS-TBNA, EBBX did not significantly add to the diagnostic yield in sarcoidosis unless visible lesions were observed.

Amyloid-related imaging abnormalities (ARIA): diagnosis, management, and care in the setting of amyloid-modifying therapy.

Amyloid-related imaging abnormalities, were originally described by dementia care experts. The wider use of aducanumab and now lecanemab warrant broader understanding by the health care provider continuum. The optimal care approach for patients with Alzheimer's dementia, treated with amyloid-targeted therapy, includes proper clinical diagnosis, complication surveillance, specific imaging protocols, expert specialty consultation, integrated treatment strategies, and proper facility system planning. Improved awareness and understanding of amyloid-modifying therapy, both benefits and potential complications, among the health care provider continuum is paramount to the success of complex care programs. Specifically, recognition of treatment high risk, high benefit groups, and the interface of concurrent antiplatelet and anticoagulation. This integrated acute, specialty, and primary care approach should improve patient care quality and outcome.

The role of infliximab in treating refractory cardiac sarcoidosis. Case series and systematic review of literature.

Cardiac sarcoidosis is associated with significant morbidity and mortality. Immunosuppressive treatment focuses on suppressing myocardial inflammation, which can lead to major adverse events especially when progressing to fibrosis. Conventional management usually includes steroids and steroid sparing agents such as methotrexate and azathioprine. Tumour necrosis factor alpha inhibitors are often reserved for those with a worsening clinical status and/or evidence of persistent inflammatory activity despite conventional therapy. Refractory cardiac sarcoidosis (CS) can be defined as the persistence or progression of active disease, evidenced either by lack of clinical response or persistence or progression of imaging abnormalities, despite being on conventional therapy. In the United Kingdom, tumour necrosis factor alpha inhibitors are currently not licensed for cardiac sarcoidosis as there are no randomised controlled trials to assess the efficacy of infliximab in this patient cohort. In this study, we present the outcomes of six patients treated with infliximab for refractory cardiac sarcoidosis at Royal Brompton Hospital and performed a systematic review of the existing literature on use of infliximab in cardiac sarcoidosis. We searched the Cochrane Library, OVID Medline, OVID Embase, Web of Science and Pubmed to identify 7 full-text studies assessing the role of infliximab in the management of cardiac sarcoidosis. Infliximab was found to play a vital role in stabilising refractory cardiac sarcoidosis by stemming clinical deterioration, arrythmia burden and even reducing steroids requirements. Further prospective trial data is necessary to validate these findings.

Imaging abnormalities of the pancreas in diabetes: implications for diagnosis and treatment.

Radiographic imaging of the pancreas has drawn recent interest as pancreas volume may serve as a biomarker in identifying the likelihood of diabetes development, subtyping diabetes, and identifying prognostic indicators of poor ultimate outcomes. In this review, the role of pancreas imaging is discussed in various forms of diabetes including type 1 diabetes (T1D), type 2 diabetes (T2D), and diabetes of the exocrine pancreas, particularly diabetes following acute or chronic pancreatitis. Recent literature of quantitative pancreatic imaging correlating with various forms of diabetes was reviewed. Imaging-derived pancreas volumes are lower in individuals with diabetes, in particular those with T1D. Additionally, morphologic changes, enhancement characteristics, fat content, and MRI signal changes have been observed in different diabetes subtypes. These characteristics, as well as potential confounding variables, are reviewed. Additionally, future areas of research in MRI, CT radiomics, and pancreatitis-related imaging predictors of diabetes are discussed. Increased understanding of pancreas imaging features which predict diabetes and gauge prognosis has the potential to identify at-risk individuals and will become increasingly important in diabetes care. This article reviews the current knowledge of common pancreas imaging features as well as future directions of ongoing research in diabetes imaging.

Thoracic high resolution computed tomography evaluation of imaging abnormalities of 108 lung cancer patients with different pulmonary function

PurposePreserved ratio impaired spirometry (PRISm) and chronic obstructive pulmonary disease (COPD) belong to lung function injury. PRISm is a precursor to COPD. We compared and evaluated the different basic information, imaging findings and survival curves of 108 lung cancer patients with different pulmonary function based on high resolution computed tomography (HRCT).MethodsThis retrospective study was performed on 108 lung cancer patients who did pulmonary function test (PFT) and thoracic HRCT. The basic information was evaluated: gender, age, body mass index (BMI), smoke, smoking index (SI). The following pulmonary function findings were evaluated: forced expiratory volume in 1s (FEV1), forced vital capacity (FVC), FEV1/FVC ratio. The following computed tomography (CT) findings were evaluated: appearance (bronchiectasis, pneumonectasis, atelectasis, ground-glass opacities [GGO], interstitial inflammation, thickened bronchial wall), diameter (aortic diameter, pulmonary artery diameter, MPAD/AD ratio, inferior vena cava diameter [IVCD]), tumor (volume, classification, distribution, staging [I, II, III, IV]). Mortality rates were calculated and survival curves were estimated using the Kaplan-Meier method.ResultsCompared with normal pulmonary function group, PRISm group and COPD group were predominantly male, older, smoked more, poorer lung function and had shorter survival time after diagnosis. There were more abnormal images in PRISm group and COPD group than in normal lung function group (N-C group). In PRISm group and COPD group, lung cancer was found late, and the tumor volume was larger, mainly central squamous carcinoma. But the opposite was true for the N-C group. The PRISm group and COPD group had significant poor survival probability compared with the normal lung function group.ConclusionsConsiderable differences regarding basic information, pulmonary function, imaging findings and survival curves are found between normal lung function group and lung function injury group. Lung function injury (PRISm and COPD) should be taken into account in future lung cancer screening studies.

SOCR‐YOLO: Small Objects Detection Algorithm in Medical Images

ABSTRACTIn the field of medical image analysis, object detection plays a crucial role by providing interpretable diagnostic information to healthcare professionals. Although current object detection models have achieved remarkable success in conventional images, their performance in detecting abnormalities in medical images has not been as satisfactory. This is primarily due to the complexity of anatomical structures in medical images, and the fact that some lesions may have subtle features, particularly in the case of early‐stage, small‐scale abnormalities. To address this challenge, we introduce SOCR‐YOLO, a novel lesion detection model with online convolutional reparameterization based on channel shuffling. First, it employs the SOCR (Shuffled Channel with Online Convolutional Re‐parameterization) module to establish a connection between feature concatenation and computational efficiency, aiming to extract more comprehensive information while reducing time consumption. Second, it incorporates the Bi‐FPN structure to achieve multiscale feature fusion. Lastly, the loss function has been optimized to improve the model training process. We evaluated two datasets, chest x‐ray (Vindr‐CXR) and brain tumor (Br35H), provided by the Kaggle competition. Experimental results show that the proposed method has outperformed several state‐of‐the‐art models, including YOLOv8, YOLO‐NAS, and RT‐DETR, in both speed and accuracy. Notably, in the context of chest x‐ray anomaly detection, SOCR‐YOLO exhibits a 1.8% enhancement in accuracy over YOLOv8 while simultaneously reducing floating‐point operations by 26.3%. Additionally, a similar 1.8% improvement in accuracy is observed in the detection of brain tumors. The results indicate the superior ability of our model to detect multiscale variations and small lesions.

Development and validation of a smartphone-based deep-learning-enabled system to detect middle-ear conditions in otoscopic images

Middle-ear conditions are common causes of primary care visits, hearing impairment, and inappropriate antibiotic use. Deep learning (DL) may assist clinicians in interpreting otoscopic images. This study included patients over 5 years old from an ambulatory ENT practice in Strasbourg, France, between 2013 and 2020. Digital otoscopic images were obtained using a smartphone-attached otoscope (Smart Scope, Karl Storz, Germany) and labeled by a senior ENT specialist across 11 diagnostic classes (reference standard). An Inception-v2 DL model was trained using 41,664 otoscopic images, and its diagnostic accuracy was evaluated by calculating class-specific estimates of sensitivity and specificity. The model was then incorporated into a smartphone app called i-Nside. The DL model was evaluated on a validation set of 3,962 images and a held-out test set comprising 326 images. On the validation set, all class-specific estimates of sensitivity and specificity exceeded 98%. On the test set, the DL model achieved a sensitivity of 99.0% (95% confidence interval: 94.5–100) and a specificity of 95.2% (91.5–97.6) for the binary classification of normal vs. abnormal images; wax plugs were detected with a sensitivity of 100% (94.6–100) and specificity of 97.7% (95.0–99.1); other class-specific estimates of sensitivity and specificity ranged from 33.3% to 92.3% and 96.0% to 100%, respectively. We present an end-to-end DL-enabled system able to achieve expert-level diagnostic accuracy for identifying normal tympanic aspects and wax plugs within digital otoscopic images. However, the system’s performance varied for other middle-ear conditions. Further prospective validation is necessary before wider clinical deployment.