Lung and Pancreatic Tumor Characterization in the Deep Learning Era: Novel Supervised and Unsupervised Learning Approaches

Abstract

Risk stratification (characterization) of tumors from radiology images can be more accurate and faster with computeraided diagnosis (CAD) tools. Tumor characterization through such tools can also enable non-invasive cancer staging, prognosis, and foster personalized treatment planning as a part of precision medicine. In this study, we propose both supervised and unsupervised machine learning strategies to improve tumor characterization. Our first approach is based on supervised learning for which we demonstrate significant gains with deep learning algorithms, particularly by utilizing a 3D Convolutional Neural Network and Transfer Learning. Motivated by the radiologists' interpretations of the scans, we then show how to incorporate task dependent feature representations into a CAD system via a graph-regularized sparse Multi-Task Learning (MTL) framework.In the second approach, we explore an unsupervised learning algorithm to address the limited availability of labeled training data, a common problem in medical imaging applications. Inspired by learning from label proportion (LLP) approaches in computer vision, we propose to use proportion-SVM for characterizing tumors. We also seek the answer to the fundamental question about the goodness of "deep features" for unsupervised tumor classification. We evaluate our proposed supervised and unsupervised learning algorithms on two different tumor diagnosis challenges: lung and pancreas with 1018 CT and 171 MRI scans, respectively, and obtain the state-of-the-art sensitivity and specificity results in both problems.

Quality assurance of computer-aided detection and diagnosis in colonoscopy

“…To alleviate big data problems in medical imaging, DL researchers often use 2 strategies: (1) transfer learning, and (2) data augmentation. In transfer learning 59,60 , the new DL model is updated from a pretrained network model which is obtained from other fields such as computer vision ImageNET where millions of natural images are made available with precise labels to train a typical neural network. In data augmentation, on the other hand, new data are artificially generated by using the available data with certain realistic manipulations such as rotating, translating, adding noise, flipping, etc.…”

Artificial Intelligence Approach for Target Classification: A State of The Art

“…); these are curves that approximate the data. Naïve Bayes' algorithm; which gives the probability of the prediction, in knowledge of previous events [77][78][79][80][81][82]. Clustering is always using mathematics; we will group the data into packets so that in each packet the data is as close as possible to each other [83][84][85].…”

Improving the Performance of Lung Cancer Prediction Using Machine Learning Techniques on Big Data

“…They trust that this evaluation will deliver all the medical research societies with the necessary knowledge to master the concept of CNN so as to utilize it for improving the overall human healthcare system. Hussein et al [4] proposed supervised learning using 3D Convolutional neural network (3D CNN) on lung nodules data set as well as unsupervised learning SVM approach to classify benign and malignant data with a accuracy of 91%.…”