A Transfer Learning Approach Based on Ultrasound Images for Liver Cancer Detection

The convolutional neural network(CNN)is one of the main algorithms that is applied to deep transfer learning for classifying two essential types of liver lesions;Hemangioma and hepatocellular carcinoma(HCC).Ultrasound images,which are commonly available and have low cost and low risk compared to computerized tomography(CT)scan images,will be used as input for the model.A total of 350 ultrasound images belonging to 59 patients are used.The number of images with HCC is 202 and 148,respectively.These images were collected from ultrasound cases.info(28 Hemangiomas patients and 11 HCC patients),the department of radiology,the University of Washington(7 HCC patients),the Atlas of ultrasound Germany(3 HCC patients),and Radiopedia and others(10 HCC patients).The ultrasound images are divided into 225,52,and 73 for training,validation,and testing.A data augmentation technique is used to enhance the validation performance.We proposed an approach based on ensembles of the best-selected deep transfer models from the on-the-shelf models:VGG16,VGG19,DenseNet,Inception,InceptionResNet,ResNet,and EfficientNet.After tuning both the feature extraction and the classification layers,the best models are selected.Validation accuracy is used for model tuning and selection.The accuracy,sensitivity,specificity and AUROC are used to evaluate the performance.The experiments are concluded in five stages.The first stage aims to evaluate the base model performance by training the on-the-shelf models.The best accu-racy obtained in the first stage is 83.5%.In the second stage,we augmented the data and retrained the on-the-shelf models with the augmented data.The best accuracy we obtained in the second stage was 86.3%.In the third stage,we tuned the feature extraction layers of the on-the-shelf models.The best accuracy obtained in the third stage is 89%.In the fourth stage,we fine-tuned the classification layer and obtained an accuracy of 93%as the best accuracy.In the fifth stage,we applied the ensemble approach using the best three-performing models and obtained an accuracy,specificity,sensitivity,and AUROC of 94%,93.7%,95.1%,and 0.944,respectively.

Predicting the Need for ICU Admission in COVID-19 Patients Using XGBoost

It is important to determine early on which patients require ICU admissions in managing COVID-19 especially when medical resources are limited.Delay in ICU admissions is associated with negative outcomes such as mortality and cost.Therefore,early identification of patients with a high risk of respiratory failure can prevent complications,enhance risk stratification,and improve the outcomes of severely-ill hospitalized patients.In this paper,we develop a model that uses the characteristics and information collected at the time of patients’admissions and during their early period of hospitalization to accurately predict whether they will need ICU admissions.We use the data explained and organized in a window-based manner by the Sírio-Libanês hospital team(published on Kaggle).Preprocessing is applied,including imputation,cleaning,and feature selection.In the cleaning process,we remove zero-variance,redundant,and/or highly correlated(measured by the Pearson correlation coefficient)features.We use Extreme Gradient Boosting(XGBoost)with early stopping as a predictor in our developed model.We run the experiment in four stages starting from the features of Window 1 in Stage 1 and then incrementally add the features of Windows 2–4 in Stages 2–4 respectively.We achieve AUCs of 0.73,0.92,0.95,and 0.97 in those four stages.