1. LUNA-16 [Link] [Download]

2. National Lung Screening Trial - NLST [Link] [Link]

3. Lung Image Database Consortium image collection - LIDC-IDRI: [Link]

4. Public Lung Image Database - ELCAP: [Link]

5. Lung CT Challenge - SPIE-AAPM: [Link]

6. Cancer Imaging Archive: [Link] 

3D Nodule Detection

1. Free-Response Receiver Operating Characteristic (FROC) is the official evaluation metric of the LUNA16 dataset [30], which is defined as the average recall rate at 0.125, 0.25, 0.5, 1, 2, 4, and 8 false positives per scan. 

   1. True positive when nodule candidate is located within a distance R from the center of any nodules in the reference standard, where R denotes the radius of the reference nodule. 

   2. False positives: Nodule candidates is not located in the range of any reference nodules. 

2. FROCIoU: defines the true positives if the 3D Intersection over Union of nodule candidates and any reference nodules is higher than one threshold (3D IoU threshold is defined as 0.25 in experiments).

3. 3D mean Average Precision (mAP) as the detection evaluation metric. 

   1. AP@0.25 (AP at 3D IoU = 0.25).

   2. AP@0.35 (AP at 3D IoU = 0.35).

   3. APs (AP for small nodules that correspond size 0-5 mm: volume < 512).

   4. APm (AP for medium nodules that correspond size 5-10 mm: 512 < volume < 4096).

   5. APl (AP for large nodules that correspond size > 10 mm: volume > 4096). 

Paper List

• C0_AnaXNet: Anatomy Aware Multi-label Finding Classification in Chest X-ray_MICCAI_2021 [Paper]

• Evaluate the Malignancy of Pulmonary Nodules Using the 3D Deep Leaky Noisy-or Network [Paper]

• B. C. Kim, J. S. Yoon, J. S. Choi, and H. I. Suk, "Multi-scale gradual integration CNN for false positive reduction in pulmonary nodule detection", in Neural Networks, Volume 115, Pages 1-10, 2019 [Code] [Dl_Bme_LgNd_FpRedu_3dCnn_Multiscale_Src_Nn2019]

• L. Sun, Z. Wang, H. Pu, G. Yuan, L. Guo, T. Pu, and Z. Peng, "Attention-embedded complementary-stream CNN for false positive reduction in pulmonary nodule detection", in Computers in Biology and Medicine, Volume 133, 104357, 2021. [Dl_Bme_LgNd_FpRedu_Attent_Cbm2021]

• Z. Wu, R. Ge, Go. Shi, L. Zhang, Y. Chen, L. Luo, Y. Cao and H. Yu, "MD-NDNet: a multi-dimensional convolutional neural network for false-positive reduction in pulmonary nodule detection", in Physics in Medicine & Biology, Volume 65, Number 23, 2020. [Dl_Bme_LungNd_Classify_3dMdNdNet_FpRedu_ChSpAttent_Luna_Pmb2020]

Reading List

• Attention U-Net: Learning Where to Look for the Pancreas [Paper]

• TOOD: Task-aligned One-stage Object Detection_arXiv_2021 [Paper]

• Deep Learning for Lung Cancer Detection: Tackling the Kaggle Data Science Bowl 2017 Challenge [Paper]

• The Effect of the Loss on Generalization: Empirical Study on Synthetic Lung Nodule Data [Paper]

• Cross-modal Attention for MRI and Ultrasound Volume Registration [Paper] [Code]

To-do List

Review

• T_C13_Lung Nodule Detection from Feature Engineering to Deep Learning in Thoracic CT Images: a Comprehensive Review_JDI_2020 [Paper]

• T_C4_NROI based feature learning for automated tumor stage classification of pulmonary lung nodules using deep convolutional neural networks_JKSU_2019 [Paper]

• C610_An overview of deep learning in medical imaging focusing on MRI_ZMP_2019 [Paper]

2D Detection

• T_C12_DeepLN: A framework for automatic lung nodule detection using multi-resolution CT screening images_KBS_2020 [Paper]

• T_C23_NoduleNet: Decoupled False Positive Reduction for Pulmonary Nodule Detection and Segmentation_MICCAI_2019 [Paper]

• T_C14_Lung nodule detection and classification from Thorax CT-scan using RetinaNet with transfer learning_JKSU_2020 [Paper]

• C4_Towards radiologist-level cancer risk assessment in CT lung screening using deep learning_CMIG_2021 [Paper]

3D Detection

• C518_End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography_Nature_2019 (NLST Data + 3D Detection + Full-patient Risk Classification) [Paper]

Transformer

• Detection of tuberculosis from chest X-ray images: Boosting the performance with vision transformer and transfer learning_ESA_2021 [Paper]

• Multi-image Super Resolution of Remotely Sensed Images using Residual Feature Attention Deep Neural Networks [Paper]

Semi-supervised

• Semi-supervised multi-task learning for lung cancer diagnosis [Paper]

• A new semi-supervised self-training method for lung cancer prediction [Paper]

Pneumonia Detection => other Deseases

• [AugmentRot + Multitask] Deep Learning for Automatic Pneumonia Detection_CVPR_2020 [Paper]

• [Dilated + Stacked] CovXNet: A multi-dilation convolutional neural network for automaticCOVID-19 and other pneumonia detection from chest X-ray images withtransferable multi-receptive feature optimization [Paper]

Lung Cancer Diagnose

• [Capsule Network + Multitask] Encoding Visual Attributes in Capsules for Explainable Medical Diagnoses [Paper]

• [Nodule Augment 3D] Synthetic Lung Nodule 3D Image Generation Using Autoencoders [Paper]

• [Patch-based + Heatmap] Knowledge-based Analysis for Mortality Prediction from CT Images [Paper]

Lung Nodule Detection [Link]

• [Capsule Network + Multitask] Encodin

Unfilter

• nnDetection: A Self-configuring Method for Medical Object Detection_2021

• C77_Models Genesis: Generic Autodidactic Models for 3D Medical Image Analysis_MICCAI 2019 [Paper] [Code - TensorFlow] 

• C7_Transferable Visual Words: Exploiting the Semantics of Anatomical Patterns for Self-supervised Learning_IEEE TMI 2021 [Paper] [Code - TensorFlow]

• Learning Visual Context by Comparison

• UNet++: A Nested U-Net Architecture for Medical Image Segmentation_2018

• COVID-19 Detection from X-ray Images using Multi-Kernel-Size Spatial-Channel Attention Network_PR_2021

• Efficient Medical Image Segmentation Based on Knowledge Distillation_2021 [Paper]

• Toward accurate MRI bone and cartilage segmentation in small data sets via an improved mask RCNN: data from the osteoarthritis initiative [Paper]

• CT male pelvic organ segmentation using fully convolutional networks with boundary sensitive representation [Paper]

• Alleviation of COVID by means of Social Distancing & Face Mask Detection Using YOLO V4 [Paper]

Other Materials

• Vietnam Cooperation for Lung health: [Link]

To Tony

• [Tool]: 

  1. https://github.com/MIC-DKFZ/nnDetection

  2. nnDetection: A Self-configuring Method for Medical Object Detection

• [Detection by Multi-view]: Learning Visual Context by Comparison.

• [Attention on Medical Image]: COVID-19 Detection from X-ray Images using Multi-Kernel-Size Spatial-Channel Attention Network

• [3D Unet + Attention]: RA-UNet: A hybrid deep attention-aware network to extract liver and tumor in CT scans

• [For FP-reduction]: ProCAN: Progressive Growing Channel Attentive NonLocal Network for Lung Nodule Classification 

• [Self-Supervised (Later)]: 

  1. FocalMix: Semi-Supervised Learning for 3D Medical Image Detection 

  2. Efficient Medical Image Segmentation Based on Knowledge Distillation

  3. Self-training with Noisy Student improves ImageNet classification

  4. [Vietnam] Utilizing Knowledge Distillation in Deep Learning for Classification of Chest X-Ray Abnormalities