Deep Volumetric Universal Lesion Detection Using Light-Weight Pseudo 3D Convolution and Surface Point Regression

Cai, Jinzheng; Yan, Ke; Cheng, Chi‐Tung; Xiao, Jing; Liao, Chien‐Hung; Lü, Le; Harrison, Adam P.

doi:10.1007/978-3-030-59719-1_1

Cited by 27 publications

(36 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Both models could reach the highest sensitivity scores for the average false positives per image greater than False Positive (FP)@2. Furthermore, both models hang slightly behind their counterparts, with dilation of [2,2,2] or [1,2,4], in the midsection between FP@0.25 and FP@2. Additionally, the models with dilation of [2,4,8] reach the highest sensitivity score of the lowest FP@0.125 while their counterparts, with dilation of [4,4,4], stay slightly behind (0.3−0.5 %).…”

Section: B Resultsmentioning

confidence: 92%

“…Replacing the last convolutional layers of Conv3-5 blocks with MDC layers within the backbone was tested with unsatisfactory results. However, the dilation and padding combinations used for the experiments are [1, 1, 1], [2, 2, 2], [4,4,4], [8,8,8], [1,2,4], and [2,4,8], where the number in brackets represents the dilation and padding for each of the three convolutional layers, respectively. The adjusted padding must keep the generated feature maps' size simultaneously throughout the different models.…”

Section: A Settingsmentioning

confidence: 99%

“…Recently, Universal Lesion Detection (ULD), as an essential technology of CADe systems, has shown effectiveness in many research studies [25], [2], [26], [22], [30]. The ULD focuses on detecting lesions throughout the whole body.…”

Section: Introductionmentioning

confidence: 99%

“…Moreover, dilated convolutions are combined with the modulated deformable convolutions to fine-tune the VGG-16 model with multi-scale receptive fields. In the experiments evaluated on the DeepLesion dataset, the modulated deformable positivesensitive region of interest pooling model achieves the highest sensitivity score of 58.8 % on average with dilation of [4,4,4] and outperforms state-of-the-art models in the range of [2,8] average false positives per image. This research demonstrates the suitability of dilation modifications and the possibility of enhancing the performance using a modulated deformable positive-sensitive region of interest pooling layer for universal lesion detectors.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Deformable Dilated Faster R-CNN for Universal Lesion Detection in CT Images

Hellmann

Ren

André

et al. 2021

2021 43rd Annual International Conference of the IEEE Engineering in Medicine &Amp; Biology Society (EMBC)

View full text Add to dashboard Cite

Cancer is a major public health issue and takes the second-highest toll of deaths caused by non-communicable diseases worldwide. Automatically detecting lesions at an early stage is essential to increase the chance of a cure. This study proposes a novel dilated Faster R-CNN with modulated deformable convolution and modulated deformable positive-sensitive region of interest pooling to detect lesions in computer tomography images. A pre-trained VGG-16 is transferred as the backbone of Faster R-CNN, followed by a region proposal network and a region of interest pooling layer to achieve lesion detection. The modulated deformable convolutional layers are employed to learn deformable convolutional filters, while the modulated deformable positive-sensitive region of interest pooling provides an enhanced feature extraction on the feature maps. Moreover, dilated convolutions are combined with the modulated deformable convolutions to fine-tune the VGG-16 model with multi-scale receptive fields. In the experiments evaluated on the DeepLesion dataset, the modulated deformable positivesensitive region of interest pooling model achieves the highest sensitivity score of 58.8 % on average with dilation of [4,4,4] and outperforms state-of-the-art models in the range of [2,8] average false positives per image. This research demonstrates the suitability of dilation modifications and the possibility of enhancing the performance using a modulated deformable positive-sensitive region of interest pooling layer for universal lesion detectors.

show abstract

Section: B Resultsmentioning

confidence: 92%

Section: A Settingsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Deformable Dilated Faster R-CNN for Universal Lesion Detection in CT Images

Hellmann

Ren

André

et al. 2021

2021 43rd Annual International Conference of the IEEE Engineering in Medicine &Amp; Biology Society (EMBC)

View full text Add to dashboard Cite

show abstract

“…Based on convolutional neural network (CNN), we developed a universal lesion detector (ULDor) that can help radiologists find all potential lesions within one unified computing framework [11] and trained it on six public lesion datasets including DeepLesion dataset. The model has shown good performance in previous internal validation [10][11][12][13][14] but the generalization needs external validation. The study aims to evaluate the model's ability to generalize with unseen data.…”

Section: Introductionmentioning

confidence: 96%

External and Internal Validation of a Computer Assisted Diagnostic Model for Detecting Multi-Organ Mass Lesions in CT images

Yan²,

Lü³

et al. 2021

Chinese Medical Sciences Journal

View full text Add to dashboard Cite

Keywords：lesion detection;| Computer-Aided Diagnosis CAD; convolutional neural network; deep learning Objective：We developed a universal lesion detector (ULDor) which showed good performance in inlab experiments. The study aims to evaluate the performance and its ability to generalize in clinical setting via both external and internal validation.Methods：The ULDor system consists of a convolutional neural network (CNN) trained on around 80K lesion annotations from about 12K CT studies in the DeepLesion dataset and 5 other public organspecific datasets. During the validation process, the test sets include two parts: the external validation datasets were comprised of 164 sets of non-contrasted chest-upper abdomen CT scans collected from a comprehensive hospital, and the internal validation datasets are comprised of 187 sets of low-dose helical CT scans from the National Lung Screening Trial (NLST). We run the model on the two test sets to output lesion detection. Three board-certified radiologists read the CT scans and verify the detection results by ULDor. We used positive predictive value (PPV) and sensitivity to measure the performance of the model in detecting space-occupying lesions at all extra-pulmonary organs, including the liver, kidney, pancreas, thyroid, lymph nodes, body wall, and thoracic spine at visualized CT images.Results：In the external validation, the lesion-level PPV and sensitivity of the model are 58% and 66%, respectively. On average, the model detected 2.1 findings per set, and among them, 0.9 were false positives. ULDor worked well for detecting liver lesions, with a PPV of 79% and a sensitivity of 93%, followed by kidney, with a PPV of 70% and a sensitivity of 58%. In internal validation with NLST test set, ULDor obtained a PPV of 75% and a sensitivity of 52% despite the relatively high noise level of soft tissue in CT images.

show abstract

A flexible three‐dimensional heterophase computed tomography hepatocellular carcinoma detection algorithm for generalizable and practical screening

Cheng

Cai²,

Teng

et al. 2022

Hepatol Commun

Self Cite

View full text Add to dashboard Cite

Hepatocellular carcinoma (HCC) can be potentially discovered from abdominal computed tomography (CT) studies under varied clinical scenarios (e.g., fully dynamic contrast-enhanced [DCE] studies, noncontrast [NC] plus venous phase [VP] abdominal studies, or NC-only studies). Each scenario presents its own clinical challenges that could benefit from computer-aided detection (CADe) tools. We investigate whether a single CADe model can be made flexible enough to handle different contrast protocols and whether this flexibility imparts performance gains. We developed a flexible three-dimensional deep algorithm, called heterophase volumetric detection (HPVD), that can accept any combination of contrast-phase inputs with adjustable sensitivity depending on the clinical purpose. We trained HPVD on 771 DCE CT scans to detect HCCs and evaluated it on 164 positives and 206 controls. We compared performance against six clinical readers, including two radiologists, two hepatopancreaticobiliary surgeons, and two hepatologists. The area under the curve of the localization receiver operating characteristic for NC-only, NC plus VP, and full DCE CT yielded 0.71 (95% confidence interval [CI], 0.64-0.77), 0.81 (95% CI, 0.75-0.87), and 0.89 (95% CI, 0.84-0.93), respectively. At a high-sensitivity operating point of 80% on DCE CT, HPVD achieved 97% specificity, which is comparable to measured physician performance. We also demonstrated performance improvements over more typical and less flexible nonheterophase detectors. Conclusion: A single deep-learning algorithm can be effectively applied to diverse HCC detection clinical scenarios, indicating that HPVD could serve as a useful clinical aid for at-risk and opportunistic HCC surveillance.

show abstract

Deep Volumetric Universal Lesion Detection Using Light-Weight Pseudo 3D Convolution and Surface Point Regression

Cited by 27 publications

References 22 publications

Deformable Dilated Faster R-CNN for Universal Lesion Detection in CT Images

Deformable Dilated Faster R-CNN for Universal Lesion Detection in CT Images

External and Internal Validation of a Computer Assisted Diagnostic Model for Detecting Multi-Organ Mass Lesions in CT images

A flexible three‐dimensional heterophase computed tomography hepatocellular carcinoma detection algorithm for generalizable and practical screening

Contact Info

Product

Resources

About