Automatic Tumor Segmentation via False Positive Reduction Network for
Whole-Body Multi-Modal PET/CT Images
- URL: http://arxiv.org/abs/2209.07705v1
- Date: Fri, 16 Sep 2022 04:01:14 GMT
- Title: Automatic Tumor Segmentation via False Positive Reduction Network for
Whole-Body Multi-Modal PET/CT Images
- Authors: Yige Peng, Jinman Kim, Dagan Feng, Lei Bi
- Abstract summary: In PET/CT image assessment, automatic tumor segmentation is an important step.
Existing methods tend to over-segment the tumor regions and include regions such as the normal high organs, inflammation, and other infections.
We introduce a false positive reduction network to overcome this limitation.
- Score: 12.885308856495353
- License: http://creativecommons.org/licenses/by-nc-sa/4.0/
- Abstract: Multi-modality Fluorodeoxyglucose (FDG) positron emission tomography /
computed tomography (PET/CT) has been routinely used in the assessment of
common cancers, such as lung cancer, lymphoma, and melanoma. This is mainly
attributed to the fact that PET/CT combines the high sensitivity for tumor
detection of PET and anatomical information from CT. In PET/CT image
assessment, automatic tumor segmentation is an important step, and in recent
years, deep learning based methods have become the state-of-the-art.
Unfortunately, existing methods tend to over-segment the tumor regions and
include regions such as the normal high uptake organs, inflammation, and other
infections. In this study, we introduce a false positive reduction network to
overcome this limitation. We firstly introduced a self-supervised pre-trained
global segmentation module to coarsely delineate the candidate tumor regions
using a self-supervised pre-trained encoder. The candidate tumor regions were
then refined by removing false positives via a local refinement module. Our
experiments with the MICCAI 2022 Automated Lesion Segmentation in Whole-Body
FDG-PET/CT (AutoPET) challenge dataset showed that our method achieved a dice
score of 0.9324 with the preliminary testing data and was ranked 1st place in
dice on the leaderboard. Our method was also ranked in the top 7 methods on the
final testing data, the final ranking will be announced during the 2022 MICCAI
AutoPET workshop. Our code is available at:
https://github.com/YigePeng/AutoPET_False_Positive_Reduction.
Related papers
- Lung-CADex: Fully automatic Zero-Shot Detection and Classification of Lung Nodules in Thoracic CT Images [45.29301790646322]
Computer-aided diagnosis can help with early lung nodul detection and facilitate subsequent nodule characterization.
We propose CADe, for segmenting lung nodules in a zero-shot manner using a variant of the Segment Anything Model called MedSAM.
We also propose, CADx, a method for the nodule characterization as benign/malignant by making a gallery of radiomic features and aligning image-feature pairs through contrastive learning.
arXiv Detail & Related papers (2024-07-02T19:30:25Z) - Revolutionizing Disease Diagnosis with simultaneous functional PET/MR and Deeply Integrated Brain Metabolic, Hemodynamic, and Perfusion Networks [40.986069119392944]
We propose MX-ARM, a multimodal MiXture-of-experts Alignment Reconstruction and Model.
It is modality detachable and exchangeable, allocating different multi-layer perceptrons dynamically ("mixture of experts") through learnable weights to learn respective representations from different modalities.
arXiv Detail & Related papers (2024-03-29T08:47:49Z) - Contrastive Diffusion Model with Auxiliary Guidance for Coarse-to-Fine
PET Reconstruction [62.29541106695824]
This paper presents a coarse-to-fine PET reconstruction framework that consists of a coarse prediction module (CPM) and an iterative refinement module (IRM)
By delegating most of the computational overhead to the CPM, the overall sampling speed of our method can be significantly improved.
Two additional strategies, i.e., an auxiliary guidance strategy and a contrastive diffusion strategy, are proposed and integrated into the reconstruction process.
arXiv Detail & Related papers (2023-08-20T04:10:36Z) - CancerUniT: Towards a Single Unified Model for Effective Detection,
Segmentation, and Diagnosis of Eight Major Cancers Using a Large Collection
of CT Scans [45.83431075462771]
Human readers or radiologists routinely perform full-body multi-organ multi-disease detection and diagnosis in clinical practice.
Most medical AI systems are built to focus on single organs with a narrow list of a few diseases.
CancerUniT is a query-based Mask Transformer model with the output of multi-tumor prediction.
arXiv Detail & Related papers (2023-01-28T20:09:34Z) - Exploring Vanilla U-Net for Lesion Segmentation from Whole-body
FDG-PET/CT Scans [16.93163630413171]
Since FDG-PET scans only provide metabolic information, healthy tissue or benign disease with irregular glucose consumption may be mistaken for cancer.
In this paper, we explore the potential of U-Net for lesion segmentation in whole-body FDG-PET/CT scans from three aspects, including network architecture, data preprocessing, and data augmentation.
Our method achieves first place in both preliminary and final leaderboards of the autoPET 2022 challenge.
arXiv Detail & Related papers (2022-10-14T03:37:18Z) - PriorNet: lesion segmentation in PET-CT including prior tumor appearance
information [0.0]
We propose a two-step approach to improve the segmentation performances of tumoral lesions in PET-CT images.
The first step generates a prior tumor appearance map from the PET-CT volumes, regarded as prior tumor information.
The second step, consisting of a standard U-Net, receives the prior tumor appearance map and PET-CT images to generate the lesion mask.
arXiv Detail & Related papers (2022-10-05T12:31:42Z) - AutoPET Challenge 2022: Automatic Segmentation of Whole-body Tumor
Lesion Based on Deep Learning and FDG PET/CT [0.0]
We propose a novel training strategy to build deep learning models capable of systemic tumor segmentation.
Our method is validated on the training set of the AutoPET 2022 Challenge.
arXiv Detail & Related papers (2022-08-31T09:14:44Z) - Lung-Originated Tumor Segmentation from Computed Tomography Scan (LOTUS)
Benchmark [48.30502612686276]
Lung cancer is one of the deadliest cancers, and its effective diagnosis and treatment depend on the accurate delineation of the tumor.
Human-centered segmentation, which is currently the most common approach, is subject to inter-observer variability.
The 2018 VIP Cup started with a global engagement from 42 countries to access the competition data.
In a nutshell, all the algorithms proposed during the competition, are based on deep learning models combined with a false positive reduction technique.
arXiv Detail & Related papers (2022-01-03T03:06:38Z) - Multi-Scale Input Strategies for Medulloblastoma Tumor Classification
using Deep Transfer Learning [59.30734371401316]
Medulloblastoma is the most common malignant brain cancer among children.
CNN has shown promising results for MB subtype classification.
We study the impact of tile size and input strategy.
arXiv Detail & Related papers (2021-09-14T09:42:37Z) - Deep PET/CT fusion with Dempster-Shafer theory for lymphoma segmentation [17.623576885481747]
Lymphoma detection and segmentation from PET/CT volumes are crucial for surgical indication and radiotherapy.
We propose an lymphoma segmentation model using an UNet with an evidential PET/CT fusion layer.
Our method get accurate segmentation results with Dice score of 0.726, without any user interaction.
arXiv Detail & Related papers (2021-08-11T19:24:40Z) - Multimodal Spatial Attention Module for Targeting Multimodal PET-CT Lung
Tumor Segmentation [11.622615048002567]
Multimodal spatial attention module (MSAM) learns to emphasize regions related to tumors.
MSAM can be applied to common backbone architectures and trained end-to-end.
arXiv Detail & Related papers (2020-07-29T10:27:22Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.