TMTV-Net: fully automated total metabolic tumor volume segmentation in lymphoma PET/CT images - a multi-center generalizability analysis.

PURPOSE: Total metabolic tumor volume (TMTV) segmentation has significant value enabling quantitative imaging biomarkers for lymphoma management. In this work, we tackle the challenging task of automated tumor delineation in lymphoma from PET/CT scans using a cascaded approach. METHODS: Our study included 1418 2-[18F]FDG PET/CT scans from four different centers. The dataset was divided into 900 scans for development/validation/testing phases and 518 for multi-center external testing. The former consisted of 450 lymphoma, lung cancer, and melanoma scans, along with 450 negative scans, while the latter consisted of lymphoma patients from different centers with diffuse large B cell, primary mediastinal large B cell, and classic Hodgkin lymphoma cases. Our approach involves resampling PET/CT images into different voxel sizes in the first step, followed by training multi-resolution 3D U-Nets on each resampled dataset using a fivefold cross-validation scheme. The models trained on different data splits were ensemble. After applying soft voting to the predicted masks, in the second step, we input the probability-averaged predictions, along with the input imaging data, into another 3D U-Net. Models were trained with semi-supervised loss. We additionally considered the effectiveness of using test time augmentation (TTA) to improve the segmentation performance after training. In addition to quantitative analysis including Dice score (DSC) and TMTV comparisons, the qualitative evaluation was also conducted by nuclear medicine physicians. RESULTS: Our cascaded soft-voting guided approach resulted in performance with an average DSC of 0.68 ± 0.12 for the internal test data from developmental dataset, and an average DSC of 0.66 ± 0.18 on the multi-site external data (n = 518), significantly outperforming (p < 0.001) state-of-the-art (SOTA) approaches including nnU-Net and SWIN UNETR. While TTA yielded enhanced performance gains for some of the comparator methods, its impact on our cascaded approach was found to be negligible (DSC: 0.66 ± 0.16). Our approach reliably quantified TMTV, with a correlation of 0.89 with the ground truth (p < 0.001). Furthermore, in terms of visual assessment, concordance between quantitative evaluations and clinician feedback was observed in the majority of cases. The average relative error (ARE) and the absolute error (AE) in TMTV prediction on external multi-centric dataset were ARE = 0.43 ± 0.54 and AE = 157.32 ± 378.12 (mL) for all the external test data (n = 518), and ARE = 0.30 ± 0.22 and AE = 82.05 ± 99.78 (mL) when the 10% outliers (n = 53) were excluded. CONCLUSION: TMTV-Net demonstrates strong performance and generalizability in TMTV segmentation across multi-site external datasets, encompassing various lymphoma subtypes. A negligible reduction of 2% in overall performance during testing on external data highlights robust model generalizability across different centers and cancer types, likely attributable to its training with resampled inputs. Our model is publicly available, allowing easy multi-site evaluation and generalizability analysis on datasets from different institutions.

Spatial process decomposition for quantitative imaging biomarkers using multiple images of varying shapes.

Quantitative imaging biomarkers (QIB) are extracted from medical images in radiomics for a variety of purposes including noninvasive disease detection, cancer monitoring, and precision medicine. The existing methods for QIB extraction tend to be ad hoc and not reproducible. In this article, a general and flexible statistical approach is proposed for handling up to three-dimensional medical images and reasonably capturing features with respect to specific spatial patterns. In particular, a model-based spatial process decomposition is developed where the random weights are unique to individual patients for component functions common across patients. Model fitting and selection are based on maximum likelihood, while feature extractions are via optimal prediction of the underlying true image. Simulation studies are conducted to investigate the properties of the proposed methodology. For illustration, a cancer image data set is analyzed and QIBs are extracted in association with a clinical endpoint.

ConTEXTual Net: A Multimodal Vision-Language Model for Segmentation of Pneumothorax.

Radiology narrative reports often describe characteristics of a patient's disease, including its location, size, and shape. Motivated by the recent success of multimodal learning, we hypothesized that this descriptive text could guide medical image analysis algorithms. We proposed a novel vision-language model, ConTEXTual Net, for the task of pneumothorax segmentation on chest radiographs. ConTEXTual Net extracts language features from physician-generated free-form radiology reports using a pre-trained language model. We then introduced cross-attention between the language features and the intermediate embeddings of an encoder-decoder convolutional neural network to enable language guidance for image analysis. ConTEXTual Net was trained on the CANDID-PTX dataset consisting of 3196 positive cases of pneumothorax with segmentation annotations from 6 different physicians as well as clinical radiology reports. Using cross-validation, ConTEXTual Net achieved a Dice score of 0.716±0.016, which was similar to the degree of inter-reader variability (0.712±0.044) computed on a subset of the data. It outperformed vision-only models (Swin UNETR: 0.670±0.015, ResNet50 U-Net: 0.677±0.015, GLoRIA: 0.686±0.014, and nnUNet 0.694±0.016) and a competing vision-language model (LAVT: 0.706±0.009). Ablation studies confirmed that it was the text information that led to the performance gains. Additionally, we show that certain augmentation methods degraded ConTEXTual Net's segmentation performance by breaking the image-text concordance. We also evaluated the effects of using different language models and activation functions in the cross-attention module, highlighting the efficacy of our chosen architectural design.

Voxel-Level Dosimetry for Combined Iodine 131 Radiopharmaceutical Therapy and External Beam Radiation Therapy Treatment Paradigms for Head and Neck Cancer.

PURPOSE: Targeted radiopharmaceutical therapy (RPT) in combination with external beam radiation therapy (EBRT) shows promise as a method to increase tumor control and mitigate potential high-grade toxicities associated with re-treatment for patients with recurrent head and neck cancer. This work establishes a patient-specific dosimetry framework that combines Monte Carlo-based dosimetry from the 2 radiation modalities at the voxel level using deformable image registration (DIR) and radiobiological constructs for patients enrolled in a phase 1 clinical trial combining EBRT and RPT. METHODS AND MATERIALS: Serial single-photon emission computed tomography (SPECT)/computed tomography (CT) patient scans were performed at approximately 24, 48, 72, and 168 hours postinjection of 577.2 MBq/m2 (15.6 mCi/m2) CLR 131, an iodine 131-containing RPT agent. Using RayStation, clinical EBRT treatment plans were created with a treatment planning CT (TPCT). SPECT/CT images were deformably registered to the TPCT using the Elastix DIR module in 3D Slicer software and assessed by measuring mean activity concentrations and absorbed doses. Monte Carlo EBRT dosimetry was computed using EGSnrc. RPT dosimetry was conducted using RAPID, a GEANT4-based RPT dosimetry platform. Radiobiological metrics (biologically effective dose and equivalent dose in 2-Gy fractions) were used to combine the 2 radiation modalities. RESULTS: The DIR method provided good agreement for the activity concentrations and calculated absorbed dose in the tumor volumes for the SPECT/CT and TPCT images, with a maximum mean absorbed dose difference of -11.2%. Based on the RPT absorbed dose calculations, 2 to 4 EBRT fractions were removed from patient EBRT treatments. For the combined treatment, the absorbed dose to target volumes ranged from 57.14 to 75.02 Gy. When partial volume corrections were included, the mean equivalent dose in 2-Gy fractions to the planning target volume from EBRT + RPT differed -3.11% to 1.40% compared with EBRT alone. CONCLUSIONS: This work demonstrates the clinical feasibility of performing combined EBRT + RPT dosimetry on TPCT scans. Dosimetry guides treatment decisions for EBRT, and this work provides a bridge for the same paradigm to be implemented within the rapidly emerging clinical RPT space.

Personalized Impression Generation for PET Reports Using Large Language Models.

Large language models (LLMs) have shown promise in accelerating radiology reporting by summarizing clinical findings into impressions. However, automatic impression generation for whole-body PET reports presents unique challenges and has received little attention. Our study aimed to evaluate whether LLMs can create clinically useful impressions for PET reporting. To this end, we fine-tuned twelve open-source language models on a corpus of 37,370 retrospective PET reports collected from our institution. All models were trained using the teacher-forcing algorithm, with the report findings and patient information as input and the original clinical impressions as reference. An extra input token encoded the reading physician's identity, allowing models to learn physician-specific reporting styles. To compare the performances of different models, we computed various automatic evaluation metrics and benchmarked them against physician preferences, ultimately selecting PEGASUS as the top LLM. To evaluate its clinical utility, three nuclear medicine physicians assessed the PEGASUS-generated impressions and original clinical impressions across 6 quality dimensions (3-point scales) and an overall utility score (5-point scale). Each physician reviewed 12 of their own reports and 12 reports from other physicians. When physicians assessed LLM impressions generated in their own style, 89% were considered clinically acceptable, with a mean utility score of 4.08/5. On average, physicians rated these personalized impressions as comparable in overall utility to the impressions dictated by other physicians (4.03, P = 0.41). In summary, our study demonstrated that personalized impressions generated by PEGASUS were clinically useful in most cases, highlighting its potential to expedite PET reporting by automatically drafting impressions.

Automatic Quantification of Serial PET/CT Images for Pediatric Hodgkin Lymphoma Patients Using a Longitudinally-Aware Segmentation Network.

Purpose: Automatic quantification of longitudinal changes in PET scans for lymphoma patients has proven challenging, as residual disease in interim-therapy scans is often subtle and difficult to detect. Our goal was to develop a longitudinally-aware segmentation network (LAS-Net) that can quantify serial PET/CT images for pediatric Hodgkin lymphoma patients. Materials and Methods: This retrospective study included baseline (PET1) and interim (PET2) PET/CT images from 297 patients enrolled in two Children's Oncology Group clinical trials (AHOD1331 and AHOD0831). LAS-Net incorporates longitudinal cross-attention, allowing relevant features from PET1 to inform the analysis of PET2. Model performance was evaluated using Dice coefficients for PET1 and detection F1 scores for PET2. Additionally, we extracted and compared quantitative PET metrics, including metabolic tumor volume (MTV) and total lesion glycolysis (TLG) in PET1, as well as qPET and ΔSUVmax in PET2, against physician measurements. We quantified their agreement using Spearman's ρ correlations and employed bootstrap resampling for statistical analysis. Results: LAS-Net detected residual lymphoma in PET2 with an F1 score of 0.606 (precision/recall: 0.615/0.600), outperforming all comparator methods (P<0.01). For baseline segmentation, LAS-Net achieved a mean Dice score of 0.772. In PET quantification, LAS-Net's measurements of qPET, ΔSUVmax, MTV and TLG were strongly correlated with physician measurements, with Spearman's ρ of 0.78, 0.80, 0.93 and 0.96, respectively. The performance remained high, with a slight decrease, in an external testing cohort. Conclusion: LAS-Net achieved high performance in quantifying PET metrics across serial scans, highlighting the value of longitudinal awareness in evaluating multi-time-point imaging datasets.

A Guide to Cross-Validation for Artificial Intelligence in Medical Imaging.

Artificial intelligence (AI) is being increasingly used to automate and improve technologies within the field of medical imaging. A critical step in the development of an AI algorithm is estimating its prediction error through cross-validation (CV). The use of CV can help prevent overoptimism in AI algorithms and can mitigate certain biases associated with hyperparameter tuning and algorithm selection. This article introduces the principles of CV and provides a practical guide on the use of CV for AI algorithm development in medical imaging. Different CV techniques are described, as well as their advantages and disadvantages under different scenarios. Common pitfalls in prediction error estimation and guidance on how to avoid them are also discussed. Keywords: Education, Research Design, Technical Aspects, Statistics, Supervised Learning, Convolutional Neural Network (CNN) Supplemental material is available for this article. © RSNA, 2023.

Domain-adapted Large Language Models for Classifying Nuclear Medicine Reports.

Purpose: To evaluate the impact of domain adaptation on the performance of language models in predicting five-point Deauville scores on the basis of clinical fluorine 18 fluorodeoxyglucose PET/CT reports. Materials and Methods: The authors retrospectively retrieved 4542 text reports and images for fluorodeoxyglucose PET/CT lymphoma examinations from 2008 to 2018 in the University of Wisconsin-Madison institutional clinical imaging database. Of these total reports, 1664 had Deauville scores that were extracted from the reports and served as training labels. The bidirectional encoder representations from transformers (BERT) model and initialized BERT models BioClinicalBERT, RadBERT, and RoBERTa were adapted to the nuclear medicine domain by pretraining using masked language modeling. These domain-adapted models were then compared with the non-domain-adapted versions on the task of five-point Deauville score prediction. The language models were compared against vision models, multimodal vision-language models, and a nuclear medicine physician, with sevenfold Monte Carlo cross-validation. Means and SDs for accuracy are reported, with P values from paired t testing. Results: Domain adaptation improved the performance of all language models (P = .01). For example, BERT improved from 61.3% ± 2.9 (SD) five-class accuracy to 65.7% ± 2.2 (P = .01) following domain adaptation. Domain-adapted RoBERTa (named DA RoBERTa) performed best, achieving 77.4% ± 3.4 five-class accuracy; this model performed similarly to its multimodal counterpart (named Multimodal DA RoBERTa) (77.2% ± 3.2) and outperformed the best vision-only model (48.1% ± 3.5, P ≤ .001). A physician given the task on a subset of the data had a five-class accuracy of 66%. Conclusion: Domain adaptation improved the performance of large language models in predicting Deauville scores in PET/CT reports.Keywords Lymphoma, PET, PET/CT, Transfer Learning, Unsupervised Learning, Convolutional Neural Network (CNN), Nuclear Medicine, Deauville, Natural Language Processing, Multimodal Learning, Artificial Intelligence, Machine Learning, Language Modeling Supplemental material is available for this article. © RSNA, 2023See also the commentary by Abajian in this issue.

Creation of waterproof, TLD probes for dose measurements to validate image-based radiopharmaceutical therapy dosimetry workflow.

Voxel-level dosimetry based on nuclear medicine images offers patient-specific personalization of radiopharmaceutical therapy (RPT) treatments. Clinical evidence is emerging demonstrating improvements in treatment precision in patients when voxel-level dosimetry is used compared to MIRD. Voxel-level dosimetry requires absolute quantification of activity concentrations in the patient, but images from SPECT/CT scanners are not quantitative and require calibration using nuclear medicine phantoms. While phantom studies can validate a scanner's ability to recover activity concentrations, these studies provide only a surrogate for the true metric of interest: absorbed doses. Measurements using thermoluminescent dosimeters (TLDs) are a versatile and accurate method of measuring absorbed dose. In this work, a TLD probe was manufactured that can fit into currently available nuclear medicine phantoms for the measurement of absorbed dose of RPT agents. Next, 748 MBq of I-131 was administered to a 16 ml hollow source sphere placed in a 6.4 L Jaszczak phantom in addition to six TLD probes, each holding 4 TLD-100 1 × 1 × 1 mm TLD-100 (LiF:Mg,Ti) microcubes. The phantom then underwent a SPECT/CT scan in accordance with a standard SPECT/CT imaging protocol for I-131. The SPECT/CT images were then input into a Monte Carlo based RPT dosimetry platform named RAPID and a three dimensional dose distribution in the phantom was estimated. Additionally, a GEANT4 benchmarking scenario (denoted 'idealized') was created using a stylized representation of the phantom. There was good agreement for all six probes, the differences between measurement and RAPID ranged between -5.5% and 0.9%. The difference between the measured and the idealized GEANT4 scenario was calculated and ranged from -4.3% and -20.5%. This work demonstrates good agreement between TLD measurements and RAPID. In addition, it introduces a novel TLD probe that can be easily introduced into clinical nuclear medicine workflows to provide QA of image-based dosimetry for RPT treatments.

Automatic Personalized Impression Generation for PET Reports Using Large Language Models.

Purpose: To determine if fine-tuned large language models (LLMs) can generate accurate, personalized impressions for whole-body PET reports. Materials and Methods: Twelve language models were trained on a corpus of PET reports using the teacher-forcing algorithm, with the report findings as input and the clinical impressions as reference. An extra input token encodes the reading physician's identity, allowing models to learn physician-specific reporting styles. Our corpus comprised 37,370 retrospective PET reports collected from our institution between 2010 and 2022. To identify the best LLM, 30 evaluation metrics were benchmarked against quality scores from two nuclear medicine (NM) physicians, with the most aligned metrics selecting the model for expert evaluation. In a subset of data, model-generated impressions and original clinical impressions were assessed by three NM physicians according to 6 quality dimensions (3-point scale) and an overall utility score (5-point scale). Each physician reviewed 12 of their own reports and 12 reports from other physicians. Bootstrap resampling was used for statistical analysis. Results: Of all evaluation metrics, domain-adapted BARTScore and PEGASUSScore showed the highest Spearman's ρ correlations (ρ=0.568 and 0.563) with physician preferences. Based on these metrics, the fine-tuned PEGASUS model was selected as the top LLM. When physicians reviewed PEGASUS-generated impressions in their own style, 89% were considered clinically acceptable, with a mean utility score of 4.08 out of 5. Physicians rated these personalized impressions as comparable in overall utility to the impressions dictated by other physicians (4.03, P=0.41). Conclusion: Personalized impressions generated by PEGASUS were clinically useful, highlighting its potential to expedite PET reporting.

Validation of Monte Carlo 131 I radiopharmaceutical dosimetry workflow using a 3D-printed anthropomorphic head and neck phantom.

PURPOSE: Approximately 50% of head and neck cancer (HNC) patients will experience loco-regional disease recurrence following initial courses of therapy. Retreatment with external beam radiotherapy (EBRT) is technically challenging and may be associated with a significant risk of irreversible damage to normal tissues. Radiopharmaceutical therapy (RPT) is a potential method to treat recurrent HNC in conjunction with EBRT. Phantoms are used to calibrate and add quantification to nuclear medicine images, and anthropomorphic phantoms can account for both the geometrical and material composition of the head and neck. In this study, we present the creation of an anthropomorphic, head and neck, nuclear medicine phantom, and its characterization for the validation of a Monte Carlo, SPECT image-based, 131 I RPT dosimetry workflow. METHODS: 3D-printing techniques were used to create the anthropomorphic phantom from a patient CT dataset. Three 131 I SPECT/CT imaging studies were performed using a homogeneous, Jaszczak, and an anthropomorphic phantom to quantify the SPECT images using a GE Optima NM/CT 640 with a high energy general purpose collimator. The impact of collimator detector response (CDR) modeling and volume-based partial volume corrections (PVCs) upon the absorbed dose was calculated using an image-based, Geant4 Monte Carlo RPT dosimetry workflow and compared against a ground truth scenario. Finally, uncertainties were quantified in accordance with recent EANM guidelines. RESULTS: The 3D-printed anthropomorphic phantom was an accurate re-creation of patient anatomy including bone. The extrapolated Jaszczak recovery coefficients were greater than that of the 3D-printed insert (∼22.8 ml) for both the CDR and non-CDR cases (with CDR: 0.536 vs. 0.493, non-CDR: 0.445 vs. 0.426, respectively). Utilizing Jaszczak phantom PVCs, the absorbed dose was underpredicted by 0.7% and 4.9% without and with CDR, respectively. Utilizing anthropomorphic phantom recovery coefficient overpredicted the absorbed dose by 3% both with and without CDR. All dosimetry scenarios that incorporated PVC were within the calculated uncertainty of the activity. The uncertainties in the cumulative activity ranged from 23.6% to 106.4% for Jaszczak spheres ranging in volume from 0.5 to 16 ml. CONCLUSION: The accuracy of Monte Carlo-based dosimetry for 131 I RPT in HNC was validated with an anthropomorphic phantom. In this study, it was found that Jaszczak-based PVCs were sufficient. Future applications of the phantom could involve 3D printing and characterizing patient-specific volumes for more personalized RPT dosimetry estimates.

Gadolinium-based contrast agent attenuation does not impact PET quantification in simultaneous dynamic contrast enhanced breast PET/MR.

PURPOSE: Simultaneous PET/MR imaging involves injection of a radiopharmaceutical and often also includes administration of a gadolinium-based contrast agent (GBCA). Phantom model studies indicate that attenuation of annihilation photons by GBCAs does not bias quantification metrics of PET radiopharmaceutical uptake. However, a direct comparison of attenuation-corrected PET values before and after administration of GBCA has not been performed in patients imaged with simultaneous dynamic PET/MR. The purpose of this study was to investigate the attenuating effect of GBCAs on standardized uptake value (SUV) quantification of 18 F-fluorodeoxyglucose (FDG) uptake in invasive breast cancer and normal tissues using simultaneous PET/MR. METHODS: The study included 13 women with newly diagnosed invasive breast cancer imaged using simultaneous dedicated prone breast PET/MR with FDG. PET data collection and two-point Dixon-based MR attenuation correction sequences began simultaneously before the administration of GBCA to avoid a potential impact of GBCA on the attenuation correction map. A standard clinical dose of GBCA was intravenously administered for the dynamic contrast enhanced MR sequences obtained during the simultaneous PET data acquisition. PET data were dynamically reconstructed into 60 frames of 30 s each. Three timing windows were chosen consisting of a single frame (30 s), two frames (60 s), or four frames (120 s) immediately before and after contrast administration. SUVmax and SUVmean of the biopsy-proven breast malignancy, fibroglandular tissue of the contralateral normal breast, descending aorta, and liver were calculated prior to and following GBCA administration. Percent change in the SUV metrics were calculated to test for a statistically significant, non-zero percent change using Wilcoxon signed-rank tests. RESULTS: No statistical change in SUVmax or SUVmean was found for the breast malignancies or normal anatomical regions during the timing windows before and after GBCA administration. CONCLUSIONS: GBCAs do not significantly impact the results of PET quantification by means of additional attenuation. However, GBCAs may still affect quantification by affecting MR acquisitions used for MR-based attenuation correction which this study did not address. Corrections to account for attenuation due to clinical concentrations of GBCAs are not necessary in simultaneous PET/MR examinations when MR-based attenuation correction sequences are performed prior to GBCA administration.

Nuclear Medicine and Artificial Intelligence: Best Practices for Evaluation (the RELAINCE Guidelines).

An important need exists for strategies to perform rigorous objective clinical-task-based evaluation of artificial intelligence (AI) algorithms for nuclear medicine. To address this need, we propose a 4-class framework to evaluate AI algorithms for promise, technical task-specific efficacy, clinical decision making, and postdeployment efficacy. We provide best practices to evaluate AI algorithms for each of these classes. Each class of evaluation yields a claim that provides a descriptive performance of the AI algorithm. Key best practices are tabulated as the RELAINCE (Recommendations for EvaLuation of AI for NuClear medicinE) guidelines. The report was prepared by the Society of Nuclear Medicine and Molecular Imaging AI Task Force Evaluation team, which consisted of nuclear-medicine physicians, physicists, computational imaging scientists, and representatives from industry and regulatory agencies.

Nuclear Medicine and Artificial Intelligence: Best Practices for Algorithm Development.

The nuclear medicine field has seen a rapid expansion of academic and commercial interest in developing artificial intelligence (AI) algorithms. Users and developers can avoid some of the pitfalls of AI by recognizing and following best practices in AI algorithm development. In this article, recommendations on technical best practices for developing AI algorithms in nuclear medicine are provided, beginning with general recommendations and then continuing with descriptions of how one might practice these principles for specific topics within nuclear medicine. This report was produced by the AI Task Force of the Society of Nuclear Medicine and Molecular Imaging.

Anatomy and Physiology of Artificial Intelligence in PET Imaging.

Artificial intelligence (AI) has seen an explosion in interest within nuclear medicine. This interest is driven by the rapid progress and eye-catching achievements of machine learning algorithms. The growing foothold of AI in molecular imaging is exposing nuclear medicine personnel to new technology and terminology. Clinicians and researchers can be easily overwhelmed by numerous architectures and algorithms that have been published. This article dissects the backbone of most AI algorithms: the convolutional neural network. The algorithm training workflow and the key ingredients and operations of a convolutional neural network are described in detail. Finally, the ubiquitous U-Net is explained step-by-step.

Artificial Intelligence-Based Data Corrections for Attenuation and Scatter in Position Emission Tomography and Single-Photon Emission Computed Tomography.

Recent developments in artificial intelligence (AI) technology have enabled new developments that can improve attenuation and scatter correction in PET and single-photon emission computed tomography (SPECT). These technologies will enable the use of accurate and quantitative imaging without the need to acquire a computed tomography image, greatly expanding the capability of PET/MR imaging, PET-only, and SPECT-only scanners. The use of AI to aid in scatter correction will lead to improvements in image reconstruction speed, and improve patient throughput. This article outlines the use of these new tools, surveys contemporary implementation, and discusses their limitations.

Guidelines on Setting Up Stations for Remote Viewing of Nuclear Medicine and Molecular Imaging Studies During COVID-19.

The current pandemic has created a situation where nuclear medicine practitioners and medical physicists read or process nuclear medicine images remotely from their home office. This article presents recommendations on the components and specifications when setting up a remote viewing station for nuclear medicine imaging.

Measuring Glucose Uptake in Primary Invasive Breast Cancer Using Simultaneous Time-of-Flight Breast PET/MRI: A Method Comparison Study with Prone PET/CT.

Purpose: To compare the measurement of glucose uptake in primary invasive breast cancer using simultaneous, time-of-flight breast PET/MRI with prone time-of-flight PET/CT. Materials and Methods: In this prospective study, women with biopsy-proven invasive breast cancer undergoing preoperative breast MRI from 2016 to 2018 were eligible. Participants who had fasted underwent prone PET/CT of the breasts approximately 60 minutes after injection of 370 MBq (10 mCi) fluorine 18 fluorodeoxyglucose (18F-FDG) followed by prone PET/MRI using standard clinical breast MRI sequences performed simultaneously with PET acquisition. Volumes of interest were drawn for tumors and contralateral normal breast fibroglandular tissue to calculate standardized uptake values (SUVs). Spearman correlation, Wilcoxon signed ranked test, Mann-Whitney test, and Bland-Altman analyses were performed. Results: Twenty-three women (mean age, 50 years; range, 33-70 years) were included. Correlation between tumor uptake values measured with PET/MRI and PET/CT was strong (r s = 0.95-0.98). No difference existed between modalities for tumor maximum SUV (SUVmax) normalized to normal breast tissue SUVmean (normSUVmax) (P = .58). The least amount of measurement bias was observed with normSUVmax, +3.86% (95% limits of agreement: -28.92, +36.64). Conclusion: These results demonstrate measurement agreement between PET/CT, the current reference standard for tumor glucose uptake quantification, and simultaneous time-of-flight breast 18F-FDG PET/MRI.Keywords: Breast, Comparative Studies, PET/CT, PET/MR Supplemental material is available for this article. © RSNA, 2021See also the commentary by Mankoff and Surti in this issue.

Impact of scanner harmonization on PET-based treatment response assessment in metastatic melanoma.

Patients with metastatic melanoma often receive 18F-FDG PET/CT scans on different scanners throughout their monitoring period. In this study, we quantified the impact of scanner harmonization on longitudinal changes in PET standardized uptake values using various harmonization and normalization methods, including an anthropomorphic PET phantom. Twenty metastatic melanoma patients received at least two FDG PET/CT scans, each on two different scanners with an average of 4 months (range: 2-8) between. Scans from a General Electric (GE) Discovery 710 PET CT-1 were harmonized to the GE Discovery VCT using image reconstruction settings matching recovery coefficients in an anthropomorphic phantom with bone equivalent inserts and wall-less synthetic lesions. In patient images, SUVmax was measured for each melanoma lesion and time-point. Lesions were classified as progressing, stable, or responding based on pre-defined threshold of ±30% change in SUVmax. For comparison, harmonization was also performed using simpler methods, including harmonization using a NEMA phantom, post-reconstruction filtering, reference region normalization of SUVmax, and use of SUVpeak instead of SUVmax. In the 20 patients, 90 lesions across two time-points were available for treatment response assessment. Treatment response classification changed in 47% (42/90) of cases after harmonization with anthropomorphic phantom. Before harmonization, 37% (33/90) of the lesions were classified as stable (changing less than 30% between two time-points), while the fraction of stable lesions increased to 58% (52/90) after harmonization. Harmonization with the NEMA phantom agreed with harmonization with the anthropomorphic phantom in 91% (82/90) of cases. Post-reconstruction filtering agreed with anthropomorphic phantom-based harmonization in 83% (75/90) cases. The utilization of reference regions for normalization or SUVpeak was unable to correct for changes as identified by the anthropomorphic phantom-based harmonization. Overall, PET scanner harmonization has a major impact on individual lesion treatment response classification in metastatic melanoma patients. Harmonization using the NEMA phantom yielded similar results to harmonization using anthropomorphic phantom, while the only acceptable post-reconstruction technique was post-reconstruction filtering. Phantom-based harmonization is therefore strongly recommended when comparing lesion uptake across time-points when the images have been acquired on different PET scanners.

Comparison of 11 automated PET segmentation methods in lymphoma.

Segmentation of lymphoma lesions in FDG PET/CT images is critical in both assessing individual lesions and quantifying patient disease burden. Simple thresholding methods remain common despite the large heterogeneity in lymphoma lesion location, size, and contrast. Here, we assess 11 automated PET segmentation methods for their use in two scenarios: individual lesion segmentation and patient-level disease quantification in lymphoma. Lesions on 18F-FDG PET/CT scans of 90 lymphoma patients were contoured by a nuclear medicine physician. Thresholding, active contours, clustering, adaptive region-growing, and convolutional neural network (CNN) methods were implemented on all physician-identified lesions. Lesion-level segmentation was evaluated using multiple segmentation performance metrics (Dice, Hausdorff Distance). Patient-level quantification of total disease burden (SUVtotal) and metabolic tumor volume (MTV) was assessed using Spearman's correlation coefficients between the segmentation output and physician contours. Lesion segmentation and patient quantification performance was compared to inter-physician agreement in a subset of 20 patients segmented by a second nuclear medicine physician. In total, 1223 lesions with median tumor-to-background ratio of 4.0 and volume of 1.8 cm3, were evaluated. When assessed for lesion segmentation, a 3D CNN, DeepMedic, achieved the highest performance across all evaluation metrics. DeepMedic, clustering methods, and an iterative threshold method had lesion-level segmentation performance comparable to the degree of inter-physician agreement. For patient-level SUVtotal and MTV quantification, all methods except 40% and 50% SUVmax and adaptive region-growing achieved a performance that was similar the agreement of the two physicians. Multiple methods, including a 3D CNN, clustering, and an iterative threshold method, achieved both good lesion-level segmentation and patient-level quantification performance in a population of 90 lymphoma patients. These methods are thus recommended over thresholding methods such as 40% and 50% SUVmax, which were consistently found to be significantly outside the limits defined by inter-physician agreement.