activation mapping
Recently Published Documents


TOTAL DOCUMENTS

300
(FIVE YEARS 130)

H-INDEX

28
(FIVE YEARS 5)

2022 ◽  
Vol 8 ◽  
Author(s):  
Yiping Jiao ◽  
Jie Yuan ◽  
Oluwatofunmi Modupeoluwa Sodimu ◽  
Yong Qiang ◽  
Yichen Ding

Deep neural networks have become the mainstream approach for analyzing and interpreting histology images. In this study, we established and validated an interpretable DNN model to assess endomyocardial biopsy (EMB) data of patients with myocardial injury. Deep learning models were used to extract features and classify EMB histopathological images of heart failure cases diagnosed with either ischemic cardiomyopathy or idiopathic dilated cardiomyopathy and non-failing cases (organ donors without a history of heart failure). We utilized the gradient-weighted class activation mapping (Grad-CAM) technique to emphasize injured regions, providing an entry point to assess the dominant morphology in the process of a comprehensive evaluation. To visualize clustered regions of interest (ROI), we utilized uniform manifold approximation and projection (UMAP) embedding for dimension reduction. We further implemented a multi-model ensemble mechanism to improve the quantitative metric (area under the receiver operating characteristic curve, AUC) to 0.985 and 0.992 on ROI-level and case-level, respectively, outperforming the achievement of 0.971 ± 0.017 and 0.981 ± 0.020 based on the sub-models. Collectively, this new methodology provides a robust and interpretive framework to explore local histopathological patterns, facilitating the automatic and high-throughput quantification of cardiac EMB analysis.


Author(s):  
Ippei Tsuboi ◽  
Michio Ogano ◽  
Kei Kimura ◽  
Hidekazu Kawanaka ◽  
Masaharu Tajiri ◽  
...  

Introduction: There is increasing evidence of the epicardial connection between the right-sided pulmonary vein (PV) carina and right atrium interrupts right-sided PV isolation after circumferential PV ablation in patients with atrial fibrillation. In such cases, carina ablation is often required. This study aimed to assess the utility of the right atrial posterior wall (RAPW) pacing in the detection of the right-sided epicardial connection (EC), evaluate the requirement for additional carina ablation after circumferential pulmonary vein (PV) ablation depending on the presence of EC, and investigate the clinical characteristics including the amount of epicardial adipose tissue (EAT) in patients with ECs. Methods and Results: Forty-one patients scheduled for PV isolation were enrolled. Before ablation, activation mapping of the LA was prospectively performed during pacing from the RAPW. EC was observed in 12 patients (EC group, 29%), whereas no EC was observed in the remaining 29 patients (non-EC group, 71%). For PV isolation, carina ablation was required in addition to circumferential ablation in 7 patients (58%) in the EC group, compared to 2 patients (7%) in the non-EC group (p < 0.003). Periatrial and intercaval EAT volumes were significantly lower (12.8 ± 6.2 vs. 23.1 ± 13.9 ml/m , p < 0.02, and 1.1 ± 0.8 vs. 2.2 ± 1.6 ml/m , p< 0.02, respectively) and the patients were younger (66.5 ± 6.6 vs. 72.4 ± 8.3 years, p < 0.03) in the EC group than in the non-EC group. Conclusions: RAPW pacing revealed EC between the RA and right PV carina in nearly a quarter of the patients.


Author(s):  
Diogo R. Ferreira ◽  
Tiago A. Martins ◽  
Paulo Rodrigues

Abstract In the nuclear fusion community, there are many specialized techniques to analyze the data coming from a variety of diagnostics. One of such techniques is the use of spectrograms to analyze the magnetohydrodynamic (MHD) behavior of fusion plasmas. Physicists look at the spectrogram to identify the oscillation modes of the plasma, and to study instabilities that may lead to plasma disruptions. One of the major causes of disruptions occurs when an oscillation mode interacts with the wall, stops rotating, and becomes a locked mode. In this work, we use deep learning to predict the occurrence of locked modes from MHD spectrograms. In particular, we use a Convolutional Neural Network (CNN) with Class Activation Mapping (CAM) to pinpoint the exact behavior that the model thinks is responsible for the locked mode. Surprisingly, we find that, in general, the model explanation agrees quite well with the physical interpretation of the behavior observed in the spectrogram.


Author(s):  
Cristiano de Oliveira Dietrich ◽  
Lucas de Oliveira Hollanda ◽  
Claudio Cirenza ◽  
Angelo Amato Vincenzo de Paola

Background Ventricular tachycardia (VT) in patients with chronic chagasic cardiomyopathy (CCC) is associated with considerable morbidity and mortality. Catheter ablation of VT in patients with CCC is very complex and challenging. The main goal of this work was to assess the efficacy of VT catheter ablation guided by late potentials (LPs) in patients with CCC. Methods and Results Seventeen consecutive patients with refractory VT and CCC were prospectively included in the study. Combined endo‐epicardial voltage and late activation mapping were obtained during baseline rhythm to define scarred and LP areas, respectively. The end point of the ablation procedure was the elimination of all identified LPs. Epicardial and endocardial dense scars (<0.5 mV) were detected in 17/17 and 15/17 patients, respectively. LPs were detected in the epicardial scars of 16/17 patients and in the endocardial scars of 14/15 patients. A total of 63 VTs were induced in 17 patients; 22/63 (33%) were stable and entrained, presenting LPs recorded in the isthmus sites. The end point of ablation was achieved in 15 of 17 patients. Ablation was not completed in 2 patients because of cardiac tamponade or vicinity of the phrenic nerve and circumflex artery. Three patients (2 with unsuccessful ablation) had VT recurrence during follow‐up (39 months). Conclusions Endo‐epicardial LP mapping allows us to identify the putative isthmuses of different VTs and effectively perform catheter ablation in patients with CCC and drug‐refractory VTs.


2021 ◽  
Vol 8 ◽  
Author(s):  
Fan Xu ◽  
Li Jiang ◽  
Wenjing He ◽  
Guangyi Huang ◽  
Yiyi Hong ◽  
...  

Background: Artificial intelligence (AI) has great potential to detect fungal keratitis using in vivo confocal microscopy images, but its clinical value remains unclarified. A major limitation of its clinical utility is the lack of explainability and interpretability.Methods: An explainable AI (XAI) system based on Gradient-weighted Class Activation Mapping (Grad-CAM) and Guided Grad-CAM was established. In this randomized controlled trial, nine ophthalmologists (three expert ophthalmologists, three competent ophthalmologists, and three novice ophthalmologists) read images in each of the conditions: unassisted, AI-assisted, or XAI-assisted. In unassisted condition, only the original IVCM images were shown to the readers. AI assistance comprised a histogram of model prediction probability. For XAI assistance, explanatory maps were additionally shown. The accuracy, sensitivity, and specificity were calculated against an adjudicated reference standard. Moreover, the time spent was measured.Results: Both forms of algorithmic assistance increased the accuracy and sensitivity of competent and novice ophthalmologists significantly without reducing specificity. The improvement was more pronounced in XAI-assisted condition than that in AI-assisted condition. Time spent with XAI assistance was not significantly different from that without assistance.Conclusion: AI has shown great promise in improving the accuracy of ophthalmologists. The inexperienced readers are more likely to benefit from the XAI system. With better interpretability and explainability, XAI-assistance can boost ophthalmologist performance beyond what is achievable by the reader alone or with black-box AI assistance.


2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Daesung Kang ◽  
Hye Mi Gweon ◽  
Na Lae Eun ◽  
Ji Hyun Youk ◽  
Jeong-Ah Kim ◽  
...  

AbstractThis study aimed to assess the diagnostic performance of deep convolutional neural networks (DCNNs) in classifying breast microcalcification in screening mammograms. To this end, 1579 mammographic images were collected retrospectively from patients exhibiting suspicious microcalcification in screening mammograms between July 2007 and December 2019. Five pre-trained DCNN models and an ensemble model were used to classify the microcalcifications as either malignant or benign. Approximately one million images from the ImageNet database had been used to train the five DCNN models. Herein, 1121 mammographic images were used for individual model fine-tuning, 198 for validation, and 260 for testing. Gradient-weighted class activation mapping (Grad-CAM) was used to confirm the validity of the DCNN models in highlighting the microcalcification regions most critical for determining the final class. The ensemble model yielded the best AUC (0.856). The DenseNet-201 model achieved the best sensitivity (82.47%) and negative predictive value (NPV; 86.92%). The ResNet-101 model yielded the best accuracy (81.54%), specificity (91.41%), and positive predictive value (PPV; 81.82%). The high PPV and specificity achieved by the ResNet-101 model, in particular, demonstrated the model effectiveness in microcalcification diagnosis, which, in turn, may considerably help reduce unnecessary biopsies.


Author(s):  
Ziguan Zhang ◽  
Wuyang Zheng ◽  
Dehua He ◽  
Zichao Hu ◽  
Qiang Xie ◽  
...  

2021 ◽  
Vol 2021 ◽  
pp. 1-7
Author(s):  
Peizhen Xie ◽  
Ke Zuo ◽  
Jie Liu ◽  
Mingliang Chen ◽  
Shuang Zhao ◽  
...  

At present, deep learning-based medical image diagnosis had achieved high performance in several diseases. However, the black-box nature of the convolutional neural network (CNN) limits their role in diagnosis. In this study, a novel interpretable diagnosis pipeline using the CNN model was proposed. Furthermore, a sizeable melanoma database that contains 841 digital whole-slide images (WSIs) was built to train and evaluate the model. The model achieved strong melanoma classification ability (0.962 areas under the receiver operating characteristic, 0.887 sensitivity, and 0.925 specificity). Moreover, the proposed model outperformed the existing schemes in terms of accuracy that is 20 pathologists (0.933 vs 0.732 accuracy). Finally, the gradient-weighted class activation mapping (Grad-CAM) method was used to show the inner logic of the proposed model and its feasibility to improve diagnosis process in healthcare. The mechanism of feature heat maps which is visualized through a saliency mapping has demonstrated that features learned or extracted by the proposed model are compatible with the accepted pathological features. Conclusively, the proposed model provides a rapid and accurate diagnosis by locating the distinctive features of melanoma to build doctors’ trust in the CNNs’ diagnosis results.


2021 ◽  
Author(s):  
Baptiste Lafabregue ◽  
Jonathan Weber ◽  
Pierre Gancarski ◽  
Germain Forestier

Sign in / Sign up

Export Citation Format

Share Document