Differential privacy preserved federated learning for prognostic modeling in COVID-19 patients using large multi-institutional chest CT dataset.

BACKGROUND: Notwithstanding the encouraging results of previous studies reporting on the efficiency of deep learning (DL) in COVID-19 prognostication, clinical adoption of the developed methodology still needs to be improved. To overcome this limitation, we set out to predict the prognosis of a large multi-institutional cohort of patients with COVID-19 using a DL-based model. PURPOSE: This study aimed to evaluate the performance of deep privacy-preserving federated learning (DPFL) in predicting COVID-19 outcomes using chest CT images. METHODS: After applying inclusion and exclusion criteria, 3055 patients from 19 centers, including 1599 alive and 1456 deceased, were enrolled in this study. Data from all centers were split (randomly with stratification respective to each center and class) into a training/validation set (70%/10%) and a hold-out test set (20%). For the DL model, feature extraction was performed on 2D slices, and averaging was performed at the final layer to construct a 3D model for each scan. The DensNet model was used for feature extraction. The model was developed using centralized and FL approaches. For FL, we employed DPFL approaches. Membership inference attack was also evaluated in the FL strategy. For model evaluation, different metrics were reported in the hold-out test sets. In addition, models trained in two scenarios, centralized and FL, were compared using the DeLong test for statistical differences. RESULTS: The centralized model achieved an accuracy of 0.76, while the DPFL model had an accuracy of 0.75. Both the centralized and DPFL models achieved a specificity of 0.77. The centralized model achieved a sensitivity of 0.74, while the DPFL model had a sensitivity of 0.73. A mean AUC of 0.82 and 0.81 with 95% confidence intervals of (95% CI: 0.79-0.85) and (95% CI: 0.77-0.84) were achieved by the centralized model and the DPFL model, respectively. The DeLong test did not prove statistically significant differences between the two models (p-value = 0.98). The AUC values for the inference attacks fluctuate between 0.49 and 0.51, with an average of 0.50 ± 0.003 and 95% CI for the mean AUC of 0.500 to 0.501. CONCLUSION: The performance of the proposed model was comparable to centralized models while operating on large and heterogeneous multi-institutional datasets. In addition, the model was resistant to inference attacks, ensuring the privacy of shared data during the training process.

Fully automated kidney image biomarker prediction in ultrasound scans using Fast-Unet+.

Any kidney dimension and volume variation can be a remarkable indicator of kidney disorders. Precise kidney segmentation in standard planes plays an undeniable role in predicting kidney size and volume. On the other hand, ultrasound is the modality of choice in diagnostic procedures. This paper proposes a convolutional neural network with nested layers, namely Fast-Unet++, promoting the Fast and accurate Unet model. First, the model was trained and evaluated for segmenting sagittal and axial images of the kidney. Then, the predicted masks were used to estimate the kidney image biomarkers, including its volume and dimensions (length, width, thickness, and parenchymal thickness). Finally, the proposed model was tested on a publicly available dataset with various shapes and compared with the related networks. Moreover, the network was evaluated using a set of patients who had undergone ultrasound and computed tomography. The dice metric, Jaccard coefficient, and mean absolute distance were used to evaluate the segmentation step. 0.97, 0.94, and 3.23 mm for the sagittal frame, and 0.95, 0.9, and 3.87 mm for the axial frame were achieved. The kidney dimensions and volume were evaluated using accuracy, the area under the curve, sensitivity, specificity, precision, and F1.

Automatic cardiac evaluations using a deep video object segmentation network.

BACKGROUND: Accurate cardiac volume and function assessment have valuable and significant diagnostic implications for patients suffering from ventricular dysfunction and cardiovascular disease. This study has focused on finding a reliable assistant to help physicians have more reliable and accurate cardiac measurements using a deep neural network. EchoRCNN is a semi-automated neural network for echocardiography sequence segmentation using a combination of mask region-based convolutional neural network image segmentation structure with reference-guided mask propagation video object segmentation network. RESULTS: The proposed method accurately segments the left and right ventricle regions in four-chamber view echocardiography series with a dice similarity coefficient of 94.03% and 94.97%, respectively. Further post-processing procedures on the segmented left and right ventricle regions resulted in a mean absolute error of 3.13% and 2.03% for ejection fraction and fractional area change parameters, respectively. CONCLUSION: This study has achieved excellent performance on the left and right ventricle segmentation, leading to more accurate estimations of vital cardiac parameters such as ejection fraction and fractional area change parameters in the left and right ventricle functionalities, respectively. The results represent that our method can predict an assured, accurate, and reliable cardiac function diagnosis in clinical screenings.

MFP-Unet: A novel deep learning based approach for left ventricle segmentation in echocardiography.

Segmentation of the Left ventricle (LV) is a crucial step for quantitative measurements such as area, volume, and ejection fraction. However, the automatic LV segmentation in 2D echocardiographic images is a challenging task due to ill-defined borders, and operator dependence issues (insufficient reproducibility). U-net, which is a well-known architecture in medical image segmentation, addressed this problem through an encoder-decoder path. Despite outstanding overall performance, U-net ignores the contribution of all semantic strengths in the segmentation procedure. In the present study, we have proposed a novel architecture to tackle this drawback. Feature maps in all levels of the decoder path of U-net are concatenated, their depths are equalized, and up-sampled to a fixed dimension. This stack of feature maps would be the input of the semantic segmentation layer. The performance of the proposed model was evaluated using two sets of echocardiographic images: one public dataset and one prepared dataset. The proposed network yielded significantly improved results when comparing with results from U-net, dilated U-net, Unet++, ACNN, SHG, and deeplabv3. An average Dice Metric (DM) of 0.953, Hausdorff Distance (HD) of 3.49, and Mean Absolute Distance (MAD) of 1.12 are achieved in the public dataset. The correlation graph, bland-altman analysis, and box plot showed a great agreement between automatic and manually calculated volume, area, and length.

Recognition of acute lymphoblastic leukemia cells in microscopic images using k-means clustering and support vector machine classifier.

Acute lymphoblastic leukemia is the most common form of pediatric cancer which is categorized into three L1, L2, and L3 and could be detected through screening of blood and bone marrow smears by pathologists. Due to being time-consuming and tediousness of the procedure, a computer-based system is acquired for convenient detection of Acute lymphoblastic leukemia. Microscopic images are acquired from blood and bone marrow smears of patients with Acute lymphoblastic leukemia and normal cases. After applying image preprocessing, cells nuclei are segmented by k-means algorithm. Then geometric and statistical features are extracted from nuclei and finally these cells are classified to cancerous and noncancerous cells by means of support vector machine classifier with 10-fold cross validation. These cells are also classified into their sub-types by multi-Support vector machine classifier. Classifier is evaluated by these parameters: Sensitivity, specificity, and accuracy which values for cancerous and noncancerous cells 98%, 95%, and 97%, respectively. These parameters are also used for evaluation of cell sub-types which values in mean 84.3%, 97.3%, and 95.6%, respectively. The results show that proposed algorithm could achieve an acceptable performance for the diagnosis of Acute lymphoblastic leukemia and its sub-types and can be used as an assistant diagnostic tool for pathologists.

Right ventricle functional parameters estimation in arrhythmogenic right ventricular dysplasia using a robust shape based deformable model.

Assessment of cardiac right-ventricle functions plays an essential role in diagnosis of arrhythmogenic right ventricular dysplasia (ARVD). Among clinical tests, cardiac magnetic resonance imaging (MRI) is now becoming the most valid imaging technique to diagnose ARVD. Fatty infiltration of the right ventricular free wall can be visible on cardiac MRI. Finding right-ventricle functional parameters from cardiac MRI images contains segmentation of right-ventricle in each slice of end diastole and end systole phases of cardiac cycle and calculation of end diastolic and end systolic volume and furthermore other functional parameters. The main problem of this task is the segmentation part. We used a robust method based on deformable model that uses shape information for segmentation of right-ventricle in short axis MRI images. After segmentation of right-ventricle from base to apex in end diastole and end systole phases of cardiac cycle, volume of right-ventricle in these phases calculated and then, ejection fraction calculated. We performed a quantitative evaluation of clinical cardiac parameters derived from the automatic segmentation by comparison against a manual delineation of the ventricles. The manually and automatically determined quantitative clinical parameters were statistically compared by means of linear regression. This fits a line to the data such that the root-mean-square error (RMSE) of the residuals is minimized. The results show low RMSE for Right Ventricle Ejection Fraction and Volume (≤ 0.06 for RV EF, and ≤ 10 mL for RV volume). Evaluation of segmentation results is also done by means of four statistical measures including sensitivity, specificity, similarity index and Jaccard index. The average value of similarity index is 86.87%. The Jaccard index mean value is 83.85% which shows a good accuracy of segmentation. The average of sensitivity is 93.9% and mean value of the specificity is 89.45%. These results show the reliability of proposed method in these cases that manual segmentation is inapplicable. Huge shape variety of right-ventricle led us to use a shape prior based method and this work can develop by four-dimensional processing for determining the first ventricular slices.