Universal lesion detection in CT scans using neural network ensembles

  title={Universal lesion detection in CT scans using neural network ensembles},
  author={Tarun Mattikalli and T. Mathai and Ronald M. Summers},
  booktitle={Medical Imaging},
In clinical practice, radiologists are reliant on the lesion size when distinguishing metastatic from non-metastatic lesions. A prerequisite for lesion sizing is their detection, as it promotes the downstream assessment of tumor spread. However, lesions vary in their size and appearance in CT scans, and radiologists often miss small lesions during a busy clinical day. To overcome these challenges, we propose the use of state-of-the-art detection neural networks to flag suspicious lesions… 

CLIP-Driven Universal Model for Organ Segmentation and Tumor Detection

The CLIP-Driven Universal Model allows better segmentation of 25 organs and 6 types of tumors by exploiting the semantic relationship among abdominal structures and achieves the state-of-the-art results on Beyond The Cranial Vault.



Improving RetinaNet for CT Lesion Detection with Dense Masks from Weak RECIST Labels

This work proposes a highly accurate and efficient one-stage lesion detector, by re-designing a RetinaNet to meet the particular challenges in medical imaging, and optimize the anchor configurations using a differential evolution search algorithm.

Uldor: A Universal Lesion Detector For Ct Scans With Pseudo Masks And Hard Negative Example Mining

This work builds a Universal Lesion Detector (ULDor) based on Mask R-CNN, which is able to detect all different kinds of lesions from whole body parts and proposes a hard negative example mining strategy to reduce the false positives for improving the detection performance.

MULAN: Multitask Universal Lesion Analysis Network for Joint Lesion Detection, Tagging, and Segmentation

A multitask universal lesion analysis network (MULAN) for joint detection, tagging, and segmentation of lesions in a variety of body parts, which greatly extends existing work of single-task lesions analysis on specific body parts.

Recist-Net: Lesion Detection Via Grouping Keypoints On Recist-Based Annotation

RecIST-Net is proposed, a new approach to lesion detection in which the four extreme points and center point of the RECIST diameters are detected and achieves a sensitivity of 92.49% at four false positives per image, outperforming other recent methods including those using multi-task learning.

Learning From Multiple Datasets With Heterogeneous and Partial Labels for Universal Lesion Detection in CT

A universal lesion detection algorithm to detect a variety of lesions and proposed strategies to mine missing annotations from partially-labeled datasets by exploiting clinical prior knowledge and cross-dataset knowledge transfer are proposed.

3D Context Enhanced Region-based Convolutional Neural Network for End-to-End Lesion Detection

3D context enhanced region-based CNN (3DCE) is proposed to incorporate 3D context information efficiently by aggregating feature maps of 2D images to detect lesions from computed tomography scans.

Lesion-Harvester: Iteratively Mining Unlabeled Lesions and Hard-Negative Examples at Scale

This work introduces Lesion-Harvester—a powerful system to harvest missing annotations from lesion datasets at high precision, and presents a pseudo 3D IoU evaluation metric that corresponds much better to the real 3 D IoU than current DeepLesion evaluation metrics.

Focal Loss for Dense Object Detection

This paper proposes to address the extreme foreground-background class imbalance encountered during training of dense detectors by reshaping the standard cross entropy loss such that it down-weights the loss assigned to well-classified examples, and develops a novel Focal Loss, which focuses training on a sparse set of hard examples and prevents the vast number of easy negatives from overwhelming the detector during training.

MMDetection: Open MMLab Detection Toolbox and Benchmark

This paper presents MMDetection, an object detection toolbox that contains a rich set of object detection and instance segmentation methods as well as related components and modules, and conducts a benchmarking study on different methods, components, and their hyper-parameters.