RadFormer: Transformers with global–local attention for interpretable and accurate Gallbladder Cancer detection

We propose a novel deep neural network architecture to learn interpretable representation for medical image analysis. Our architecture generates a global attention for region of interest, and then learns bag of words style deep feature embeddings with local attention. The global, and local feature maps are combined using a contemporary transformer architecture for highly accurate Gallbladder Cancer (GBC) detection from Ultrasound (USG) images. Our experiments indicate that the detection accuracy of our model beats even human radiologists, and advocates its use as the second reader for GBC diagnosis. Bag of words embeddings allow our model to be probed for generating interpretable explanations for GBC detection consistent with the ones reported in medical literature. We show that the proposed model not only helps understand decisions of neural network models but also aids in discovery of new visual features relevant to the diagnosis of GBC. Source-code is available at https://github.com/sbasu276/RadFormer.     

Multi-Imaging Characterization of Cardiac Phenotype in Different Types of Amyloidosis

BackgroundBone scintigraphy is extremely valuable when assessing patients with suspected cardiac amyloidosis (CA), but the clinical significance and associated phenotype of different degrees of cardiac uptake across different types is yet to be defined.ObjectivesThis study sought to define the phenotypes of patients with varying degrees of cardiac uptake on bone scintigraphy, across multiple types of systemic amyloidosis, using extensive characterization comprising biomarkers as well as echocardiographic and cardiac magnetic resonance (CMR) imaging.MethodsA total of 296 patients (117 with immunoglobulin light-chain amyloidosis [AL], 165 with transthyretin amyloidosis [ATTR], 7 with apolipoprotein AI amyloidosis [AApoAI], and 7 with apolipoprotein AIV amyloidosis [AApoAIV]) underwent deep characterization of their cardiac phenotype.ResultsAL patients with grade 0 myocardial radiotracer uptake spanned the spectrum of CMR findings from no CA to characteristic CA, whereas AL patients with grades 1 to 3 always produced characteristic CMR features. In ATTR, the CA burden strongly correlated with myocardial tracer uptake, except in Ser77Tyr. AApoAI presented with grade 0 or 1 and disproportionate right-sided involvement. AApoAIV always presented with grade 0 and characteristic CA. AL grade 1 patients (n = 48; 100%) had characteristic CA, whereas only ATTR grade 1 patients with Ser77Tyr had characteristic CA on CMR (n = 5; 11.4%). After exclusion of Ser77Tyr, AApoAI, and AApoAIV, CMR showing characteristic CA or an extracellular volume of >0.40 in patients with grade 0 to 1 cardiac uptake had a sensitivity and specificity of 100% for AL.ConclusionsThere is a wide variation in cardiac phenotype between different amyloidosis types across different degrees of cardiac uptake. The combination of CMR and bone scintigraphy can help to define the diagnostic differentials and the clinical phenotype in each individual patient.     

人工智能聊天机器人助力眼科和科学论文写作

人工智能聊天机器人ChatGPT是使用深度学习技术、能对自然语言输入产生类似人类反应的一种大型语言模型(LLMs)。它由OpenAI公司在2022.11开发，属于“生成预训练转换器(GPT)”模型家族的一种，目前可为公众所用。ChatGPT能够捕捉人类语言的细微差别和复杂性，生成适当的、与上下文相关的响应。它可以帮助医务人员完成各种任务，如研究、诊断、患者监护和医学教育，从确定研究课题到协助临床和实验室诊断，了解各自领域的新进展和科学写作。ChatGPT在眼科已吸引了越来越多的关注和广泛应用。然而，目前在这些任务中使用ChatGPT和其他人工智能工具仍存在一定的局限性、伦理和法律问题，如可信度、剽窃、版权侵犯和偏见。未来的研究将集中在开发新的方法来减轻这些局限性，同时发挥ChatGPT在医疗等相关方面的积极作用。     

逆转录病毒载体介导人载脂蛋白与卵磷脂胆固醇酰基转移酶基因在骨骼肌的异源性共表达

为探讨人载脂蛋白AI和卵磷脂胆固醇酰基转移酶基因在肌源性细胞中异源共表达的可能性,构建含上述基因和新霉素磷酸转移酶基因的多顺反子重组逆转录病毒载体,以此制备重组病毒颗粒并转染小鼠原代肌母细胞及C2C12肌源性细胞株。酶联免疫吸附法和免疫组织化学检测证实转染后的细胞均具有异源共表达人载脂蛋白AI与卵磷脂胆固醇酰基转移酶的能力,经G418筛选则获得稳定转化的C2C12细胞株,60天后仍能有效共表达人载脂蛋白AI与卵磷脂胆固醇酰基转移酶,聚合酶链反应法检测显示人载脂蛋白AI cDNA与IRES序列均有效整合于靶细胞基因组中,提示以重组逆转录病毒为载体对肌源性细胞进行遗传修饰,再移植回骨骼肌使之在体内长期高效表达载脂蛋白AI和卵磷脂胆固醇酰基转移酶,可能是一种值得探讨的通过促进胆固醇逆转运途径来防止或减轻高脂血症和动脉粥样硬化的方法。     

Uncertainty aware training to improve deep learning model calibration for classification of cardiac MR images

Quantifying uncertainty of predictions has been identified as one way to develop more trustworthy artificial intelligence (AI) models beyond conventional reporting of performance metrics. When considering their role in a clinical decision support setting, AI classification models should ideally avoid confident wrong predictions and maximise the confidence of correct predictions. Models that do this are said to be well calibrated with regard to confidence. However, relatively little attention has been paid to how to improve calibration when training these models, i.e. to make the training strategy uncertainty-aware. In this work we: (i) evaluate three novel uncertainty-aware training strategies with regard to a range of accuracy and calibration performance measures, comparing against two state-of-the-art approaches, (ii) quantify the data (aleatoric) and model (epistemic) uncertainty of all models and (iii) evaluate the impact of using a model calibration measure for model selection in uncertainty-aware training, in contrast to the normal accuracy-based measures. We perform our analysis using two different clinical applications: cardiac resynchronisation therapy (CRT) response prediction and coronary artery disease (CAD) diagnosis from cardiac magnetic resonance (CMR) images. The best-performing model in terms of both classification accuracy and the most common calibration measure, expected calibration error (ECE) was the Confidence Weight method, a novel approach that weights the loss of samples to explicitly penalise confident incorrect predictions. The method reduced the ECE by 17% for CRT response prediction and by 22% for CAD diagnosis when compared to a baseline classifier in which no uncertainty-aware strategy was included. In both applications, as well as reducing the ECE there was a slight increase in accuracy from 69% to 70% and 70% to 72% for CRT response prediction and CAD diagnosis respectively. However, our analysis showed a lack of consistency in terms of optimal models when using different calibration measures. This indicates the need for careful consideration of performance metrics when training and selecting models for complex high risk applications in healthcare.     

Segmentation ability map: Interpret deep features for medical image segmentation

Deep convolutional neural networks (CNNs) have been widely used for medical image segmentation. In most studies, only the output layer is exploited to compute the final segmentation results and the hidden representations of the deep learned features have not been well understood. In this paper, we propose a prototype segmentation (ProtoSeg) method to compute a binary segmentation map based on deep features. We measure the segmentation abilities of the features by computing the Dice between the feature segmentation map and ground-truth, named as the segmentation ability score (SA score for short). The corresponding SA score can quantify the segmentation abilities of deep features in different layers and units to understand the deep neural networks for segmentation. In addition, our method can provide a mean SA score which can give a performance estimation of the output on the test images without ground-truth. Finally, we use the proposed ProtoSeg method to compute the segmentation map directly on input images to further understand the segmentation ability of each input image. Results are presented on segmenting tumors in brain MRI, lesions in skin images, COVID-related abnormality in CT images, prostate segmentation in abdominal MRI, and pancreatic mass segmentation in CT images. Our method can provide new insights for interpreting and explainable AI systems for medical image segmentation. Our code is available on: https://github.com/shengfly/ProtoSeg.     

Guidelines and evaluation of clinical explainable AI in medical image analysis

Explainable artificial intelligence (XAI) is essential for enabling clinical users to get informed decision support from AI and comply with evidence-based medical practice. Applying XAI in clinical settings requires proper evaluation criteria to ensure the explanation technique is both technically sound and clinically useful, but specific support is lacking to achieve this goal. To bridge the research gap, we propose the Clinical XAI Guidelines that consist of five criteria a clinical XAI needs to be optimized for. The guidelines recommend choosing an explanation form based on Guideline 1 (G1) Understandability and G2 Clinical relevance. For the chosen explanation form, its specific XAI technique should be optimized for G3 Truthfulness, G4 Informative plausibility, and G5 Computational efficiency. Following the guidelines, we conducted a systematic evaluation on a novel problem of multi-modal medical image explanation with two clinical tasks, and proposed new evaluation metrics accordingly. Sixteen commonly-used heatmap XAI techniques were evaluated and found to be insufficient for clinical use due to their failure in G3 and G4. Our evaluation demonstrated the use of Clinical XAI Guidelines to support the design and evaluation of clinically viable XAI.     

Is ChatGPT a valid author?

This letter to the editors takes a deeper look at the validity and ethics of authorship of a recently published article in Nurse Education in Practice in which authorship was shared with a chatbox software program, ChatGPT (https://doi.org/10.1016/j.nepr.2022.103537). In particular, a closer assessment is made of the authorship of that article from the established principles of authorship as delineated by the ICMJE.     

AI聊天机器人在骨质疏松骨折术后患者延续护理中的应用

目的 探讨AI聊天机器人在骨质疏松骨折术后患者院外延续护理中的应用效果。方法 将300例骨质疏松骨折术后患者按照时间分为常规组和干预组，每组150例。常规组实施常规院外延续护理，干预组在常规组的基础上由AI聊天机器人辅助完成院外延续护理。比较两组干预后骨质疏松健康信念水平、健康自我管理能力、疼痛评分、关节功能及二次骨折发生率。结果 出院1个月、3个月、6个月、12个月干预组骨质疏松健康信念水平、健康自我管理能力评分显著高于常规组，二次骨折发生率显著低于常规组（均P<0.05）；两组干预前及干预后各时间点疼痛及Harris髋关节功能评分比较，差异无统计学意义（均P>0.05）。结论 AI聊天机器人用于骨质疏松骨折术后患者院外延续护理，可有效提高患者骨质疏松健康信念及健康自我管理能力，降低二次骨折发生率。