The RECIST criteria are used in computed tomography (CT) imaging to assess changes in tumour burden induced by cancer therapeutics throughout treatment. One of its requirements is frequent measurement of lesion diameters , which is often time consuming for clinicians. We aimed to study clinician-interactive AI, defined as deep learning models that use image annotations as input to assist in radiological measurements. Two annotation types are compared in their enhancement of predictive capabilities: mouse clicks in the tumour region, and bounding boxes surrounding lesions. The model architectures compared in this study are the U-Net, V-Net, AH-Net, and SegRes-Net. Models were trained and tested using a non-small cell lung cancer dataset from the cancer imaging archive (TCIA) consisting of CT scans and corresponding gold-standard lesion segmentations inferred from PET/CT scans. Mouse clicks and bounding boxes, representing clinician input, were artificially generated. The absolute percent error between predicted and ground truth diameters was computed for each model architecture. Bounding box annotations yielded mean absolute percent errors of 4.9 ± 2.1 %, 7.8 ± 3.4 %, 5.6 ± 2.4 % and 5.6 ± 2.3 %, respectively, whereas models using clicks annotations yielded 17.0 ± 7.9 %, 19.8 ± 9.3 %, 21.4 ± 10.9 % and 18.1 ± 7.9%. The corresponding mean dice scores across all model architectures were 0.883 ± 0.004 and 0.760 ± 0.012 for bounding box and click annotations respectively. Models were then implemented in an AI pipeline for clinical use at the BC cancer agency using the Ascinta software package; click annotations yielded qualitatively better results than bounding box annotations.
The application of computer-vision algorithms in medical imaging has increased rapidly in recent years. However, algorithm training is challenging due to limited sample sizes, lack of labeled samples, as well as privacy concerns regarding data sharing. To address these issues, we previously developed (Bergen et al. 2022) a synthetic PET dataset for Head & Neck (H&N) cancer using the temporal generative adversarial network (TGAN) architecture and evaluated its performance segmenting lesions and identifying radiomics features in synthesized images. In this work, a two-alternative forced-choice (2AFC) observer study was performed to quantitatively evaluate the ability of human observers to distinguish between real and synthesized oncological PET images. In the study eight trained readers, including two board-certified nuclear medicine physicians, read 170 real/synthetic image pairs presented as 2D-transaxial using a dedicated web app. For each image pair, the observer was asked to identify the “real” image and input their confidence level with a 5-point Likert scale. P-values were computed using the binomial test and Wilcoxon signed-rank test. A heat map was used to compare the response accuracy distribution for the signed-rank test. Response accuracy for all observers ranged from 36.2% [27.9-44.4] to 63.1% [54.8-71.3]. Six out of eight observers did not identify the real image with statistical significance, indicating that the synthetic dataset was reasonably representative of oncological PET images. Overall, this study adds validity to the realism of our simulated H&N cancer dataset, which may be implemented in the future to train AI algorithms while favoring patient confidentiality and privacy protection.
The prostate-specific membrane antigen (PSMA) is a powerful target for positron emission tomography (PET) that has opened a new era in the diagnosis and management of prostate cancer (PCa). Aiming to provide an automated diagnostic and management tool that can help detect metastatic PCa lesions in PSMA-PET images, we deployed and investigated an array of state-of-the-art deep learning-based object detection algorithms (4 categories of multi-stage, single-stage, anchor-free, and end-to-end transformer-based). The results of 17 trained networks are reported in terms of 3 metrics (precision, recall, and F1 score), showing the ability of object detection models to localize PCa metastatic lesions of different sizes and standard uptake values (SUV). Our goal is to provide a fully automated computer-aided diagnosis (CAD) tool to assist physicians in performing the diagnosis by significantly saving time and decreasing false-negative rates. A novelty of the present work is to focus on multiple rotations of maximum intensity projection (MIP) images computed on 3D volumes in the dataset, as a new investigative training framework for detection. .
The need for accurate and consistent ground truth hinders advances in supervised learning approaches for tumor segmentation especially in PET images. In this study, we revisited the effect of supervision level on two semi-supervised approaches based on Robust FCM (RFCM) and Mumford-Shah (MS) losses for unsupervised learning combined with labeled FCM (LFCM) and Dice loss respectively as the supervised loss terms ((RFCM + αLFCM) and (MS+ αDice)). We used a multi-center (BC and SM) dataset of lymphoma patients with heterogeneous characteristics. Our results revealed that when the test data are from a center with low contribution in training data, increasing the level of supervision results in lower segmentation performance. The performance drop of MS based semi-supervised approach was higher compared to FCM based that means the training of MS based approach is more dependent on supervised learning.
Automated slice classification is clinically relevant since it can be incorporated into medical image segmentation workflows as a preprocessing step that would flag slices with a higher probability of containing tumors, thereby directing physicians’ attention to the important slices. In this work, we train a ResNet-18 network to classify axial slices of lymphoma PET/CT images (collected from two institutions) depending on whether the slice intercepted a tumor (positive slice) in the 3D image or if the slice did not (negative slice). Various instances of the network were trained on 2D axial datasets created in different ways: (i) slice-level split and (ii) patient-level split; inputs of different types were used: (i) only PET slices and (ii) concatenated PET and CT slices; and different training strategies were employed: (i) center-aware (CAW) and (ii) center-agnostic (CAG). Model performances were compared using the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC), and various binary classification metrics. We observe and describe a performance overestimation in the case of slice-level split as compared to the patient-level split training. The model trained using patient-level split data with the network input containing only PET slices in the CAG training regime was the best performing/generalizing model on a majority of metrics. Our models were additionally more closely compared using the sensitivity metric on the positive slices from their respective test sets.
Segmentation of lymphoma lesions is challenging due to their varied sizes and locations in whole-body PET scans. In this work, we present a fully-automated segmentation technique using a multi-center dataset of diffuse large B-cell lymphoma (DLBCL) with heterogeneous characteristics. We utilized a dataset of [18F]FDG-PET scans (n=194) from two different imaging centers including cases with primary mediastinal large B-cell lymphoma (PMBCL) (n=104). Automated brain and bladder removal approaches were utilized as preprocessing steps, to tackle false positives caused by normal hypermetabolic uptake in these organs. Our segmentation model is a convolutional neural network (CNN), based on a 3D U-Net architecture that includes squeeze and excitation (SE) modules. Hybrid distribution, region, and boundary-based losses (Unified Focal and Mumford-Shah (MS)) were utilized that showed the best performance compared to other combinations (p<0.05). Cross-validation between different centers, DLBCL and PMBCL cases, and three random splits were applied on train/validation data. The ensemble of these six models achieved a Dice similarity coefficient (DSC) of 0.77 ± 0.08 and Hausdorff distance (HD) of 16.5 ±12.5. Our 3D U-net model with SE modules for segmentation with hybrid loss performed significantly better (p<0.05) as compared to the 3D U-Net (without SE modules) using the same loss function (Unified Focal and MS loss) (DSC= 0.64 ± 0.21 and HD= 26.3 ± 18.7). Our model can facilitate a fully automated quantification pipeline in a multi-center context that opens the possibility for routine reporting of total metabolic tumor volume (TMTV) and other metrics shown useful for the management of lymphoma.
Accurate detection and segmentation of diffuse large B-cell lymphoma (DLBCL) from PET images has important implications for estimation of total metabolic tumor volume, radiomics analysis, surgical intervention and radiotherapy. Manual segmentation of tumors in whole-body PET images is time-consuming, labor-intensive and operator-dependent. In this work, we develop and validate a fast and efficient three-step cascaded deep learning model for automated detection and segmentation of DLBCL tumors from PET images. As compared to a single end-to-end network for segmentation of tumors in whole-body PET images, our three-step model is more effective (improves 3D Dice score from 58.9% to 78.1%) since each of its specialized modules, namely the slice classifier, the tumor detector and the tumor segmentor, can be trained independently to a high degree of skill to carry out a specific task, rather than a single network with suboptimal performance on overall segmentation.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.