Multilevel Contextual 3-D CNNs for False Positive Reduction in Pulmonary Nodule Detection

Abstract

While our method is tailored for pulmonary nodule detection, the proposed framework is general and can be easily extended to many other 3-D object detection tasks from volumetric medical images, where the targeting objects have large variations and are accompanied by a number of hard mimics.

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)…”

Evaluating the performance of a deep learning‐based computer‐aided diagnosis (DL‐CAD) system for detecting and characterizing lung nodules: Comparison with the performance of double reading by radiologists

“…With the increasing improvement of CAD systems, the majority of studies have demonstrated that CAD systems could detect more nodules than radiologists, even after double reading . Moreover, in comparison with most CAD systems based on supervised machine learning algorithms, multiple studies have shown that deep learning‐based CAD systems (DL‐CAD) have superior detection rates and further reduce false positive rates . However, CAD systems are far from perfect and thus require further development to be improved.…”

Dense networks with relative location awareness for thorax disease identification

“…, one can see even for the Nodule, which has a small lesion area, the network with the bottom‐up and top‐down structure performs better, this mainly because the large receptive field of view can extract more contextual information about disease. The contextual information around a lesion will be conducive to the diagnosis of disease, because if a lesion occurs in one location, nearby tissue will also change. In addition, we realized that the two structures that with and without bottom‐up and top‐down structures have almost the same performance for Fibrosis, in theory, the network with a large receptive field of view should have a significant improvement.…”