Learning from Experts: Developing Transferable Deep Features for Patient-Level Lung Cancer Prediction

A survey on deep learning in medical image analysis

“…As summarized in Table 2, most works employ simple Anavi et al (2015) Image retrieval Combines classical features with those from pre-trained CNN for image retrieval using SVM Bar et al (2015) Pathology detection Features from a pre-trained CNN and low level features are used to detect various diseases Anavi et al (2016) Image retrieval Continuation of Anavi et al (2015), adding age and gender as features Bar et al (2016) Pathology detection Continuation of Bar et al (2015), more experiments and adding feature selection Cicero et al (2016) Pathology detection GoogLeNet CNN detects five common abnormalities, trained and validated on a large data set Tuberculosis detection Processes entire radiographs with a pre-trained fine-tuned network with 6 convolution layers Kim and Hwang (2016) Tuberculosis detection MIL framework produces heat map of suspicious regions via deconvolution Shin et al (2016a) Pathology detection CNN detects 17 diseases, large data set (7k images), recurrent networks produce short captions Rajkomar et al (2017) Frontal/lateral classification Pre-trained CNN performs frontal/lateral classification task Yang et al (2016c) Bone suppression Cascade of CNNs at increasing resolution learns bone images from gradients of radiographs Wang et al (2016a) Nodule classification Combines classical features with CNN features from pre-trained ImageNet CNN Used a standard feature extractor and a pre-trained CNN to classify detected lesions as benign peri-fissural nodules van Detects nodules with pre-trained CNN features from orthogonal patches around candidate, classified with SVM Shen et al (2015b) Three CNNs at different scales estimate nodule malignancy scores of radiologists (LIDC-IDRI data set) Chen et al (2016e) Combines features from CNN, SDAE and classical features to characterize nodules from LIDC-IDRI data set Ciompi et al (2016) Multi-stream CNN to classify nodules into subtypes: solid, part-solid, non-solid, calcified, spiculated, perifissural Dou et al (2016b) Uses 3D CNN around nodule candidates; ranks #1 in LUNA16 nodule detection challenge Li et al (2016a) Detects nodules with 2D CNN that processes small patches around a nodule Setio et al (2016) Detects nodules with end-to-end trained multi-stream CNN with 9 patches per candidate Shen et al (2016) 3D CNN classifies volume centered on nodule as benign/malignant, results are combined to patient level prediction Sun et al (2016b) Same dataset as Shen et al (2015b)…”

Radiomics in Brain Tumor: Image Assessment, Quantitative Feature Descriptors, and Machine-Learning Approaches

“…However, the availability of labeled medical data poses a significant challenge for developing efficient deep learning models. For example, cancer images with pathologically-proven labels are costly to collect at scale, thus data integration with different type of clinical labels may present an alternative means to overcome such obstacles for deep-learning models, which need large datasets as inputs 74 . Although it remains uncertain how we gain clinical insights from these deep-learning outcomes and how we optimize network architectures for the better use of multi-scale medical data (e.g., serial MRI, genomics, and clinical data) 75 , the extraction of compact patterns via hierarchical networks presents enormous opportunities for large-scale radiomics applications (see Figure 3).…”

Integration of convolutional neural networks for pulmonary nodule malignancy assessment in a lung cancer classification pipeline

“…Despite the abundance of papers focusing on classification of nodule malignancy and on nodules detection in CT scans, little effort has been put in providing a systematic analysis of the effects of combining both to answer the question of whether predicting malignancy at a nodule level is beneficial for cancer prediction at patient level. To the best of our knowledge, [19] is the closest work to ours that tackles this question. However, the focus of [19] is limited to the transferability of deep features of nodules to the cancer prediction task and the input data are exactly located nodules.…”