scholarly journals Repeatability of two semi-automatic artificial intelligence approaches for tumor segmentation in PET

2020 ◽  
Author(s):  
Elisabeth Pfaehler ◽  
Liesbet Mesotten ◽  
Gem Kramer ◽  
Michiel Thomeer ◽  
Karolien Vanhove ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Majority Vote ◽  
Automatic Segmentation ◽  
Intraclass Correlation ◽  
Cancer Staging ◽  
Tumor Segmentation ◽  
Jaccard Coefficient ◽  
Primary Tumors ◽  
Segmentation Methods ◽  
Segmentation Approach

Abstract Background: Positron Emission Tomography (PET) is routinely used for cancer staging and treatment follow up. Metabolic active tumor volume (MATV) as well as total MATV (TMATV - including primary tumor, lymph nodes and metastasis) and/or total lesion glycolysis (TLG) derived from PET images have been identified as prognostic factor or for the evaluation of treatment efficacy in cancer patients. To this end, a segmentation approach with high precision and repeatability is important. However, the implementation of a repeatable and accurate segmentation algorithm remains an ongoing challenge. Methods: In this study, we compare two semi-automatic artificial intelligence (AI) based segmentation methods with conventional semi-automatic segmentation approaches in terms of repeatability. One segmentation approach is based on a textural feature (TF) segmentation approach designed for accurate and repeatable segmentation of primary tumors and metastasis. Moreover, a Convolutional Neural Network (CNN) is trained. The algorithms are trained, validated and tested using a lung cancer PET dataset. The segmentation accuracy of both segmentation approaches is compared using the Jaccard Coefficient (JC). Additionally, the approaches are externally tested on a fully independent test-retest dataset. The repeatability of the methods is compared with those of two majority vote (MV2, MV3) approaches, 41%SUVMAX, and a SUV>4 segmentation (SUV4). Repeatability is assessed with test-retest coefficients (TRT%) and intraclass correlation coefficient (ICC). An ICC>0.9 was regarded as representing excellent repeatability.Results: The accuracy of the segmentations with the reference segmentation was good (JC median TF: 0.7, CNN: 0.73) Both segmentation approaches outperformed most other conventional segmentation methods in terms of test-retest coefficient (TRT% mean: TF: 13.0%, CNN: 13.9%, MV2: 14.1%, MV3: 28.1%, 41%SUVMAX: 28.1%, SUV4: 18.1% ) and ICC (TF: 0.98, MV2: 0.97, CNN: 0.99, MV3: 0.73, SUV4: 0.81, and 41%SUVMAX: 0.68).Conclusion: The semi-automatic AI based segmentation approaches used in this study provided better repeatability than conventional segmentation approaches. Moreover, both algorithms lead to accurate segmentations for both primary tumors as well as metastasis and are therefore good candidates for PET tumor segmentation.

scholarly journals Repeatability of two semi-automatic artificial intelligence approaches for tumor segmentation in PET

2020 ◽  
Author(s):  
Elisabeth Pfaehler ◽  
Liesbet Mesotten ◽  
Gem Kramer ◽  
Michiel Thomeer ◽  
Karolien Vanhove ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Majority Vote ◽  
Automatic Segmentation ◽  
Intraclass Correlation ◽  
Cancer Staging ◽  
Tumor Segmentation ◽  
Jaccard Coefficient ◽  
Primary Tumors ◽  
Segmentation Methods ◽  
Segmentation Approach

Abstract Background: Positron Emission Tomography (PET) is routinely used for cancer staging and treatment follow up. Metabolic active tumor volume (MATV) as well as total MATV (TMATV - including primary tumor, lymph nodes and metastasis) and/or total lesion glycolysis (TLG) derived from PET images have been identified as prognostic factor or for the evaluation of treatment efficacy in cancer patients. To this end, a segmentation approach with high precision and repeatability is important. However, the implementation of a repeatable and accurate segmentation algorithm remains an ongoing challenge. Methods: In this study, we compare two semi-automatic artificial intelligence (AI) based segmentation methods with conventional semi-automatic segmentation approaches in terms of repeatability. One segmentation approach is based on a textural feature (TF) segmentation approach designed for accurate and repeatable segmentation of primary tumors and metastasis. Moreover, a Convolutional Neural Network (CNN) is trained. The algorithms are trained, validated and tested using a lung cancer PET dataset. The segmentation accuracy of both segmentation approaches is compared using the Jaccard Coefficient (JC). Additionally, the approaches are externally tested on a fully independent test-retest dataset. The repeatability of the methods is compared with those of two majority vote (MV2, MV3) approaches, 41%SUVMAX, and a SUV>4 segmentation (SUV4). Repeatability is assessed with test-retest coefficients (TRT%) and intraclass correlation coefficient (ICC). An ICC>0.9 was regarded as representing excellent repeatability.Results: The accuracy of the segmentations with the reference segmentation was good (JC median TF: 0.7, CNN: 0.73) Both segmentation approaches outperformed most other conventional segmentation methods in terms of test-retest coefficient (TRT% mean: TF: 13.0%, CNN: 13.9%, MV2: 14.1%, MV3: 28.1%, 41%SUVMAX: 28.1%, SUV4: 18.1% ) and ICC (TF: 0.98, MV2: 0.97, CNN: 0.99, MV3: 0.73, SUV4: 0.81, and 41%SUVMAX: 0.68).Conclusion: The semi-automatic AI based segmentation approaches used in this study provided better repeatability than conventional segmentation approaches. Moreover, both algorithms lead to accurate segmentations for both primary tumors as well as metastasis and are therefore good candidates for PET tumor segmentation.

scholarly journals Repeatability of two semi-automatic artificial intelligence approaches for tumor segmentation in PET

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Elisabeth Pfaehler ◽  
Liesbet Mesotten ◽  
Gem Kramer ◽  
Michiel Thomeer ◽  
Karolien Vanhove ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Majority Vote ◽  
Automatic Segmentation ◽  
Intraclass Correlation ◽  
Cancer Staging ◽  
Tumor Segmentation ◽  
Jaccard Coefficient ◽  
Primary Tumors ◽  
Segmentation Methods ◽  
Segmentation Approach

Abstract Background Positron emission tomography (PET) is routinely used for cancer staging and treatment follow-up. Metabolic active tumor volume (MATV) as well as total MATV (TMATV—including primary tumor, lymph nodes and metastasis) and/or total lesion glycolysis derived from PET images have been identified as prognostic factor or for the evaluation of treatment efficacy in cancer patients. To this end, a segmentation approach with high precision and repeatability is important. However, the implementation of a repeatable and accurate segmentation algorithm remains an ongoing challenge. Methods In this study, we compare two semi-automatic artificial intelligence (AI)-based segmentation methods with conventional semi-automatic segmentation approaches in terms of repeatability. One segmentation approach is based on a textural feature (TF) segmentation approach designed for accurate and repeatable segmentation of primary tumors and metastasis. Moreover, a convolutional neural network (CNN) is trained. The algorithms are trained, validated and tested using a lung cancer PET dataset. The segmentation accuracy of both segmentation approaches is compared using the Jaccard coefficient (JC). Additionally, the approaches are externally tested on a fully independent test–retest dataset. The repeatability of the methods is compared with those of two majority vote (MV2, MV3) approaches, 41%SUVMAX, and a SUV > 4 segmentation (SUV4). Repeatability is assessed with test–retest coefficients (TRT%) and intraclass correlation coefficient (ICC). An ICC > 0.9 was regarded as representing excellent repeatability. Results The accuracy of the segmentations with the reference segmentation was good (JC median TF: 0.7, CNN: 0.73). Both segmentation approaches outperformed most other conventional segmentation methods in terms of test–retest coefficient (TRT% mean: TF: 13.0%, CNN: 13.9%, MV2: 14.1%, MV3: 28.1%, 41%SUVMAX: 28.1%, SUV4: 18.1%) and ICC (TF: 0.98, MV2: 0.97, CNN: 0.99, MV3: 0.73, SUV4: 0.81, and 41%SUVMAX: 0.68). Conclusion The semi-automatic AI-based segmentation approaches used in this study provided better repeatability than conventional segmentation approaches. Moreover, both algorithms lead to accurate segmentations for both primary tumors as well as metastasis and are therefore good candidates for PET tumor segmentation.

scholarly journals Repeatability of Two Artificial Intelligence Approaches for Tumor Segmentation in PET

2020 ◽  
Author(s):  
Elisabeth Pfaehler ◽  
Liesbet Mesotten ◽  
Gem Kramer ◽  
Michiel Thomeer ◽  
Karolien Vanhove ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Majority Vote ◽  
Intraclass Correlation ◽  
Cancer Staging ◽  
Total Lesion Glycolysis ◽  
Tumor Segmentation ◽  
Jaccard Coefficient ◽  
Primary Tumors ◽  
Segmentation Methods ◽  
Segmentation Approach

Abstract Background: Positron Emission Tomography (PET) is routinely used for cancer staging and treatment follow up. Metabolic active tumor volume (MATV) as well as total MATV (TMATV - including primary tumor, lymph nodes and metastasis) and/or total lesion glycolysis (TLG) derived from PET images have been identified as prognostic factor or for the evaluation of treatment efficacy in cancer patients. To this end a segmentation approach with high precision and repeatability is important. However, the implementation of a repeatable and accurate segmentation algorithm remains an ongoing challenge. Methods: In this study, we compare two artificial intelligence (AI) based segmentation methods with conventional segmentation approaches in terms of repeatability. One segmentation approach is based on a textural feature (TF) segmentation approach designed for accurate and repeatable segmentation of primary tumors and metastasis. Moreover, a Convolutional Neural Network (CNN) is trained. The algorithms are trained, validated and tested using a lung cancer PET dataset. The segmentation accuracy of both segmentation approaches is compared using the jaccard coefficient (JC). Additionally, the approaches are applied on a fully independent test-retest dataset. The repeatability of the methods is compared with the repeatability of two majority vote (MV2, MV3) approaches, 41%SUVMAX, and a SUV>4 segmentation (SUV4). Repeatability is assessed with test-retest coefficients (TRT%) and intraclass correlation coefficient (ICC). A TRT% of 0 indicates perfect repeatability and an ICC>0.9 was regarded as representing excellent repeatability.Results: The accuracy of the segmentations with the reference segmentation was good (JC median TF: 0.7, CNN: 0.73) Both segmentation approaches outperformed together with the MV2 approach the other conventional segmentation methods in terms of test-retest coefficient (TRT% mean: TF: 13.0%, CNN: 13.9%, MV2: 14.1%, MV3: 28.1%, 41%SUVMAX: 28.1%, SUV4: 18.1% ) and ICC (TF: 0.98, MV2: 0.97, CNN: 0.99, MV3: 0.73, SUV4: 0.81, and 41%SUVMAX: 0.68).Conclusion: The AI based segmentation approaches used in this study provided better repeatability than conventional segmentation approaches. Moreover, both algorithms lead to accurate segmentations for both primary tumors as well as metastasis and are therefore good candidates for PET tumor segmentation.

An artificial intelligence framework for automatic segmentation and volumetry of vestibular schwannomas from contrast-enhanced T1-weighted and high-resolution T2-weighted MRI

2019 ◽  
pp. 1-9 ◽  
Author(s):  
Jonathan Shapey ◽  
Guotai Wang ◽  
Reuben Dorent ◽  
Alexis Dimitriadis ◽  
Wenqi Li ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
High Resolution ◽  
Tumor Volume ◽  
Automatic Segmentation ◽  
Vestibular Schwannomas ◽  
Tumor Segmentation ◽  
Imaging Data ◽  
Data Types ◽  
Automated Method ◽  
Contrast Enhanced

OBJECTIVEAutomatic segmentation of vestibular schwannomas (VSs) from MRI could significantly improve clinical workflow and assist in patient management. Accurate tumor segmentation and volumetric measurements provide the best indicators to detect subtle VS growth, but current techniques are labor intensive and dedicated software is not readily available within the clinical setting. The authors aim to develop a novel artificial intelligence (AI) framework to be embedded in the clinical routine for automatic delineation and volumetry of VS.METHODSImaging data (contrast-enhanced T1-weighted [ceT1] and high-resolution T2-weighted [hrT2] MR images) from all patients meeting the study’s inclusion/exclusion criteria who had a single sporadic VS treated with Gamma Knife stereotactic radiosurgery were used to create a model. The authors developed a novel AI framework based on a 2.5D convolutional neural network (CNN) to exploit the different in-plane and through-plane resolutions encountered in standard clinical imaging protocols. They used a computational attention module to enable the CNN to focus on the small VS target and propose a supervision on the attention map for more accurate segmentation. The manually segmented target tumor volume (also tested for interobserver variability) was used as the ground truth for training and evaluation of the CNN. We quantitatively measured the Dice score, average symmetric surface distance (ASSD), and relative volume error (RVE) of the automatic segmentation results in comparison to manual segmentations to assess the model’s accuracy.RESULTSImaging data from all eligible patients (n = 243) were randomly split into 3 nonoverlapping groups for training (n = 177), hyperparameter tuning (n = 20), and testing (n = 46). Dice, ASSD, and RVE scores were measured on the testing set for the respective input data types as follows: ceT1 93.43%, 0.203 mm, 6.96%; hrT2 88.25%, 0.416 mm, 9.77%; combined ceT1/hrT2 93.68%, 0.199 mm, 7.03%. Given a margin of 5% for the Dice score, the automated method was shown to achieve statistically equivalent performance in comparison to an annotator using ceT1 images alone (p = 4e−13) and combined ceT1/hrT2 images (p = 7e−18) as inputs.CONCLUSIONSThe authors developed a robust AI framework for automatically delineating and calculating VS tumor volume and have achieved excellent results, equivalent to those achieved by an independent human annotator. This promising AI technology has the potential to improve the management of patients with VS and potentially other brain tumors.

scholarly journals AResU-Net: Attention Residual U-Net for Brain Tumor Segmentation

Symmetry ◽  
2020 ◽  
Vol 12 (5) ◽  
pp. 721 ◽  
Author(s):  
Jianxin Zhang ◽  
Xiaogang Lv ◽  
Hengbo Zhang ◽  
Bin Liu
Keyword(s):  
Brain Tumor ◽  
Automatic Segmentation ◽  
Attention Mechanism ◽  
Tumor Segmentation ◽  
Brain Tumor Segmentation ◽  
Local Responses ◽  
Segmentation Methods ◽  
Comparable Performance ◽  
Magnetic Resonance Imaging Mri ◽  
The Brain

Automatic segmentation of brain tumors from magnetic resonance imaging (MRI) is a challenging task due to the uneven, irregular and unstructured size and shape of tumors. Recently, brain tumor segmentation methods based on the symmetric U-Net architecture have achieved favorable performance. Meanwhile, the effectiveness of enhancing local responses for feature extraction and restoration has also been shown in recent works, which may encourage the better performance of the brain tumor segmentation problem. Inspired by this, we try to introduce the attention mechanism into the existing U-Net architecture to explore the effects of local important responses on this task. More specifically, we propose an end-to-end 2D brain tumor segmentation network, i.e., attention residual U-Net (AResU-Net), which simultaneously embeds attention mechanism and residual units into U-Net for the further performance improvement of brain tumor segmentation. AResU-Net adds a series of attention units among corresponding down-sampling and up-sampling processes, and it adaptively rescales features to effectively enhance local responses of down-sampling residual features utilized for the feature recovery of the following up-sampling process. We extensively evaluate AResU-Net on two MRI brain tumor segmentation benchmarks of BraTS 2017 and BraTS 2018 datasets. Experiment results illustrate that the proposed AResU-Net outperforms its baselines and achieves comparable performance with typical brain tumor segmentation methods.

scholarly journals A Brief Review on Brain Tumor Segmentation and Detection Algorithms with Issues and Challenges

2021 ◽  
Vol 10 (4) ◽  
pp. 2885-2891
Keyword(s):  
Brain Tumor ◽  
Detection System ◽  
Automatic Segmentation ◽  
Tumor Location ◽  
Medical Image Segmentation ◽  
Tumor Segmentation ◽  
Data Types ◽  
Brain Tumor Segmentation ◽  
Segmentation Approach ◽  
Clustering Optimization

Computer vision techniques and development of computer-aided tools are evolving as the areas of research for automatic segmentation of brain tumors. Some of these techniques showed good results but there is no winning technique as these approaches have often not used practically in hospitals. In these days, research on medical healthcare system [1] is an emerging area and main focused on the designing of an efficient segmentation approach with concept of Artificial Intelligence (AI) techniques for appropriate region and fast segmentation purpose. There are a lots of clustering as well as traditional segmentation approaches are available for medical images, but most of them are depended on the data types. In this paper, we presented a brief review on clustering-based medical image segmentation with their challenging factors faced by researchers [2]. Due to high success rate of AI, Deep Learning (DL) algorithms, there has been a considerable amount of brain tumor segmentation works are aimed by researcher and try to solve the exiting challenges. In this survey, various type of brain tumor segmentation and detection system are analyzed to find out the exact tumor location and faced issued by the researchers. In Addition, some challenging factors are also analyzed with various algorithms of segmentation such DL, K-means clustering, Optimization and traditional approaches.

Assessment of automatic segmentation accuracy with various point cloud density

2020 ◽  
Vol 961 (7) ◽  
pp. 47-55
Author(s):  
A.G. Yunusov ◽  
A.J. Jdeed ◽  
N.S. Begliarov ◽  
M.A. Elshewy
Keyword(s):  
Point Cloud ◽  
Laser Scanning ◽  
Reference Data ◽  
Automatic Segmentation ◽  
Geometric Accuracy ◽  
Segmentation Method ◽  
Segmentation Methods ◽  
Segmentation Accuracy ◽  
Number Of Segments ◽  
Cloud Density

Laser scanning is considered as one of the most useful and fast technologies for modelling. On the other hand, the size of scan results can vary from hundreds to several million points. As a result, the large volume of the obtained clouds leads to complication at processing the results and increases the time costs. One way to reduce the volume of a point cloud is segmentation, which reduces the amount of data from several million points to a limited number of segments. In this article, we evaluated effect on the performance, the accuracy of various segmentation methods and the geometric accuracy of the obtained models at density changes taking into account the processing time. The results of our experiment were compared with reference data in a form of comparative analysis. As a conclusion, some recommendations for choosing the best segmentation method were proposed.

scholarly journals Structured Reporting of Rectal Cancer Staging and Restaging: A Consensus Proposal

Cancers ◽  
2021 ◽  
Vol 13 (9) ◽  
pp. 2135
Author(s):  
Vincenza Granata ◽  
Damiano Caruso ◽  
Roberto Grassi ◽  
Salvatore Cappabianca ◽  
Alfonso Reginelli ◽  
...  
Keyword(s):  
Rectal Cancer ◽  
Correlation Coefficient ◽  
Clinical Decision Making ◽  
Intraclass Correlation ◽  
Clinical Decision ◽  
Cancer Staging ◽  
Structured Reporting ◽  
Quality Analysis ◽  
Promising Tool ◽  
Structured Reports

Background: Structured reporting (SR) in oncologic imaging is becoming necessary and has recently been recognized by major scientific societies. The aim of this study was to build MRI-based structured reports for rectal cancer (RC) staging and restaging in order to provide clinicians all critical tumor information. Materials and Methods: A panel of radiologist experts in abdominal imaging, called the members of the Italian Society of Medical and Interventional Radiology, was established. The modified Delphi process was used to build the SR and to assess the level of agreement in all sections. The Cronbach’s alpha (Cα) correlation coefficient was used to assess the internal consistency of each section and to measure the quality analysis according to the average inter-item correlation. The intraclass correlation coefficient (ICC) was also evaluated. Results: After the second Delphi round of the SR RC staging, the panelists’ single scores and sum of scores were 3.8 (range 2–4) and 169, and the SR RC restaging panelists’ single scores and sum of scores were 3.7 (range 2–4) and 148, respectively. The Cα correlation coefficient was 0.79 for SR staging and 0.81 for SR restaging. The ICCs for the SR RC staging and restaging were 0.78 (p < 0.01) and 0.82 (p < 0.01), respectively. The final SR version was built and included 53 items for RC staging and 50 items for RC restaging. Conclusions: The final version of the structured reports of MRI-based RC staging and restaging should be a helpful and promising tool for clinicians in managing cancer patients properly. Structured reports collect all Patient Clinical Data, Clinical Evaluations and relevant key findings of Rectal Cancer, both in staging and restaging, and can facilitate clinical decision-making.

scholarly journals The Potential of Artificial Intelligence to Detect Lymphovascular Invasion in Testicular Cancer

Cancers ◽  
2021 ◽  
Vol 13 (6) ◽  
pp. 1325
Author(s):  
Abhisek Ghosh ◽  
Korsuk Sirinukunwattana ◽  
Nasullah Khalid Alham ◽  
Lisa Browning ◽  
Richard Colling ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Testicular Cancer ◽  
Germ Cell ◽  
Lymphovascular Invasion ◽  
Automatic Segmentation ◽  
Prognostic Significance ◽  
Germ Cell Tumours ◽  
Testicular Germ Cell ◽  
Testicular Germ Cell Tumours ◽  
Whole Slide Images

Testicular cancer is the most common cancer in men aged from 15 to 34 years. Lymphovascular invasion refers to the presence of tumours within endothelial-lined lymphatic or vascular channels, and has been shown to have prognostic significance in testicular germ cell tumours. In non-seminomatous tumours, lymphovascular invasion is the most powerful prognostic factor for stage 1 disease. For the pathologist, searching multiple slides for lymphovascular invasion can be highly time-consuming. The aim of this retrospective study was to develop and assess an artificial intelligence algorithm that can identify areas suspicious for lymphovascular invasion in histological digital whole slide images. Areas of possible lymphovascular invasion were annotated in a total of 184 whole slide images of haematoxylin and eosin (H&E) stained tissue from 19 patients with testicular germ cell tumours, including a mixture of seminoma and non-seminomatous cases. Following consensus review by specialist uropathologists, we trained a deep learning classifier for automatic segmentation of areas suspicious for lymphovascular invasion. The classifier identified 34 areas within a validation set of 118 whole slide images from 10 patients, each of which was reviewed by three expert pathologists to form a majority consensus. The precision was 0.68 for areas which were considered to be appropriate to flag, and 0.56 for areas considered to be definite lymphovascular invasion. An artificial intelligence tool which highlights areas of possible lymphovascular invasion to reporting pathologists, who then make a final judgement on its presence or absence, has been demonstrated as feasible in this proof-of-concept study. Further development is required before clinical deployment.

Sign in / Sign up

Export Citation Format

Share Document