Intensity standardization of MRI prior to radiomic feature extraction for artificial intelligence research in glioma-a systematic review.

OBJECTIVES: Radiomics is a promising avenue in non-invasive characterisation of diffuse glioma. Clinical translation is hampered by lack of reproducibility across centres and difficulty in standardising image intensity in MRI datasets. The study aim was to perform a systematic review of different methods of MRI intensity standardisation prior to radiomic feature extraction. METHODS: MEDLINE, EMBASE, and SCOPUS were searched for articles meeting the following eligibility criteria: MRI radiomic studies where one method of intensity normalisation was compared with another or no normalisation, and original research concerning patients diagnosed with diffuse gliomas. Using PRISMA criteria, data were extracted from short-listed studies including number of patients, MRI sequences, validation status, radiomics software, method of segmentation, and intensity standardisation. QUADAS-2 was used for quality appraisal. RESULTS: After duplicate removal, 741 results were returned from database and reference searches and, from these, 12 papers were eligible. Due to a lack of common pre-processing and different analyses, a narrative synthesis was sought. Three different intensity standardisation techniques have been studied: histogram matching (5/12), limiting or rescaling signal intensity (8/12), and deep learning (1/12)-only two papers compared different methods. From these studies, histogram matching produced the more reliable features compared to other methods of altering MRI signal intensity. CONCLUSION: Multiple methods of intensity standardisation have been described in the literature without clear consensus. Further research that directly compares different methods of intensity standardisation on glioma MRI datasets is required. KEY POINTS: • Intensity standardisation is a key pre-processing step in the development of robust radiomic signatures to evaluate diffuse glioma. • A minority of studies compared the impact of two or more methods. • Further research is required to directly compare multiple methods of MRI intensity standardisation on glioma datasets.

Combined PET-CT and MRI for response evaluation in patients with squamous cell anal carcinoma treated with curative-intent chemoradiotherapy.

OBJECTIVES: To assess the effectiveness of fluorine-18 fluorodeoxyglucose (FDG) positron-emission tomography-computed tomography (PET-CT) and magnetic resonance imaging (MRI) for response assessment post curative-intent chemoradiotherapy (CRT) in anal squamous cell carcinoma (ASCC). METHODS: Consecutive ASCC patients treated with curative-intent CRT at a single centre between January 2018 and April 2020 were retrospectively identified. Clinical meta-data including progression-free survival (PFS) and overall survival (OS) outcomes were collated. Three radiologists evaluated PET-CT and MRI using qualitative response assessment criteria and agreed in consensus. Two-proportion z test was used to compare diagnostic performance metrics (sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy). Kaplan-Meier analysis (Mantel-Cox log-rank) was performed. RESULTS: MRI (accuracy 76%, PPV 44.8%, NPV 95.7%) and PET-CT (accuracy 69.3%, PPV 36.7%, NPV 91.1%) performance metrics were similar; when combined, there were statistically significant improvements (accuracy 94.7%, PPV 78.9%, NPV 100%). Kaplan-Meier analysis demonstrated significant differences in PFS between responders and non-responders at PET-CT (p = 0.007), MRI (p = 0.005), and consensus evaluation (p < 0.001). Cox regression analysis of PFS demonstrated a lower hazard ratio (HR) and narrower 95% confidence intervals for consensus findings (HR = 0.093, p < 0.001). Seventy-five patients, of which 52 (69.3%) were females, with median follow-up of 17.8 months (range 5-32.6) were included. Fifteen of the 75 (20%) had persistent anorectal and/or nodal disease after CRT. Three patients died, median time to death 6.2 months (range 5-18.3). CONCLUSION: Combined PET-CT and MRI response assessment post-CRT better predicts subsequent outcome than either modality alone. This could have valuable clinical benefits by guiding personalised risk-adapted patient follow-up. KEY POINTS: • MRI and PET-CT performance metrics for assessing response following chemoradiotherapy (CRT) in patients with anal squamous cell carcinoma (ASCC) were similar. • Combined MRI and PET-CT treatment response assessment 3 months after CRT in patients with ASCC was demonstrated to be superior to either modality alone. • A combined MRI and PET-CT assessment 3 months after CRT in patients with ASCC has the potential to improve accuracy and guide optimal patient management with a greater ability to predict outcome than either modality alone.

A methodological framework for AI-assisted diagnosis of active aortitis using radiomic analysis of FDG PET-CT images: Initial analysis.

BACKGROUND: The aim of this study was to explore the feasibility of assisted diagnosis of active (peri-)aortitis using radiomic imaging biomarkers derived from [18F]-Fluorodeoxyglucose Positron Emission Tomography-Computed Tomography (FDG PET-CT) images. METHODS: The aorta was manually segmented on FDG PET-CT in 50 patients with aortitis and 25 controls. Radiomic features (RF) (n = 107), including SUV (Standardized Uptake Value) metrics, were extracted from the segmented data and harmonized using the ComBat technique. Individual RFs and groups of RFs (i.e., signatures) were used as input in Machine Learning classifiers. The diagnostic utility of these classifiers was evaluated with area under the receiver operating characteristic curve (AUC) and accuracy using the clinical diagnosis as the ground truth. RESULTS: Several RFs had high accuracy, 84% to 86%, and AUC scores 0.83 to 0.97 when used individually. Radiomic signatures performed similarly, AUC 0.80 to 1.00. CONCLUSION: A methodological framework for a radiomic-based approach to support diagnosis of aortitis was outlined. Selected RFs, individually or in combination, showed similar performance to the current standard of qualitative assessment in terms of AUC for identifying active aortitis. This framework could support development of a clinical decision-making tool for a more objective and standardized assessment of aortitis.

Cell proliferation detected using [18F]FLT PET/CT as an early marker of abdominal aortic aneurysm.

BACKGROUND: Abdominal aortic aneurysm (AAA) is a focal aortic dilatation progressing towards rupture. Non-invasive AAA-associated cell proliferation biomarkers are not yet established. We investigated the feasibility of the cell proliferation radiotracer, fluorine-18-fluorothymidine ([18F]FLT) with positron emission tomography/computed tomography (PET/CT) in a progressive pre-clinical AAA model (angiotensin II, AngII infusion). METHODS AND RESULTS: Fourteen-week-old apolipoprotein E-knockout (ApoE-/-) mice received saline or AngII via osmotic mini-pumps for 14 (n = 7 and 5, respectively) or 28 (n = 3 and 4, respectively) days and underwent 90-minute dynamic [18F]FLT PET/CT. Organs were harvested from independent cohorts for gamma counting, ultrasound scanning, and western blotting. [18F]FLT uptake was significantly greater in 14- (n = 5) and 28-day (n = 3) AAA than in saline control aortae (n = 5) (P < 0.001), which reduced between days 14 and 28. Whole-organ gamma counting confirmed greater [18F]FLT uptake in 14-day AAA (n = 9) compared to saline-infused aortae (n = 4) (P < 0.05), correlating positively with aortic volume (r = 0.71, P < 0.01). Fourteen-day AAA tissue showed increased expression of thymidine kinase-1, equilibrative nucleoside transporter (ENT)-1, ENT-2, concentrative nucleoside transporter (CNT)-1, and CNT-3 than 28-day AAA and saline control tissues (n = 3 each) (all P < 0.001). CONCLUSIONS: [18F]FLT uptake is increased during the active growth phase of the AAA model compared to saline control mice and late-stage AAA.

Second-look PET-CT following an initial incomplete PET-CT response to (chemo)radiotherapy for head and neck squamous cell carcinoma.

OBJECTIVES: The limited positive predictive value of an incomplete response on PET-CT following (chemo)radiotherapy for head and neck squamous cell carcinoma (HNSCC) means that the optimal management strategy remains uncertain. The aim of the study is to assess the utility of a 'second-look' interval PET-CT. METHODS: Patients with HNSCC who were treated with (chemo)radiotherapy between 2008 and 2017 and underwent (i) baseline and (ii) response assessment PET-CT and (iii) second-look PET-CT following incomplete (positive or equivocal scan) response were included. Endpoints were conversion rate to complete response (CR) and test characteristics of the second-look PET-CT. RESULTS: Five hundred sixty-two patients with HNSCC underwent response assessment PET-CT at a median of 17 weeks post-radiotherapy. Following an incomplete response on PET-CT, 40 patients underwent a second-look PET-CT at a median of 13 weeks (range 6-25) from the first response PET-CT. Thirty-four out of 40 (85%) patients had oropharyngeal carcinoma. Twenty-four out of 40 (60%) second-look PET-CT scans converted to a complete locoregional response. The primary tumour conversion rate was 15/27 (56%) and the lymph node conversion rate was 14/19 (74%). The sensitivity, specificity, positive predictive value and negative predictive value (NPV) of the second-look PET-CT were 75%, 75%, 25% and 96% for the primary tumour and 100%, 92%, 40% and 100% for lymph nodes. There were no cases of progression following conversion to CR in the primary site or lymph nodes. CONCLUSIONS: The majority of patients who undergo a second-look PET-CT convert to a CR. The NPV of a second-look PET-CT is high, suggesting the potential to avoid surgical intervention. KEY POINTS: • PET-CT is a useful tool for response assessment following (chemo)radiotherapy for head and neck squamous cell carcinoma. • An incomplete response on PET-CT has a limited positive predictive value and optimal management is uncertain. • These data show that with a 'second-look' interval PET-CT, the majority of patients convert to a complete metabolic response. When there is doubt about clinical and radiological response, a 'second-look' PET-CT can be used to spare patients unnecessary surgical intervention.

Assessing the impact of inadequate hydration on isotope-GFR measurement.

Guidelines state that patients undergoing isotope glomerular filtration rate (GFR) tests should maintain adequate hydration, but pragmatically these tests can coincide with procedures requiring the patient not to eat or drink ('nil-by-mouth') for up to 12 hours beforehand. This study investigated the impact of a 12-hour nil-by-mouth regime on GFR measurement. Twelve healthy volunteers were recruited from our institution. Exclusion criteria included diabetes mellitus, being under 18 years of age and pregnancy. Isotope GFR measurements were carried out on these volunteers twice. One of the tests adhered strictly to the British Nuclear Medicine Society (BNMS) guidelines for GFR measurement and the other test was carried out after the volunteers had refrained from eating or drinking anything for 12 hours. The order of these tests was randomly assigned. The results show that after a nil-by-mouth regime, participants' average absolute GFR fell from 108 ml/min to 97 ml/min (p < .01), while normalised GFR fell from 97 ml/min/1.73 m2 to 88 ml/min/1.73m2 (p < .01). Serum creatinine rose from 68 mmol/L to 73 mmol/L (p < .05). There were no changes in blood pressure, serum hydration markers or bio-impedance measured fluid status. Urine analysis showed statistically significant increases in urea, creatinine and osmolality levels after the nil-by-mouth regime. The results highlight the importance of following current guidelines recommending fluid intake during the procedure. Practitioners should consider what other outpatient appointments are being scheduled concurrently with a GFR test.

Multimodality imaging with CT, MR and FDG-PET for radiotherapy target volume delineation in oropharyngeal squamous cell carcinoma.

BACKGROUND: This study aimed to quantify the variation in oropharyngeal squamous cell carcinoma gross tumour volume (GTV) delineation between CT, MR and FDG PET-CT imaging. METHODS: A prospective, single centre, pilot study was undertaken where 11 patients with locally advanced oropharyngeal cancers (2 tonsil, 9 base of tongue primaries) underwent pre-treatment, contrast enhanced, FDG PET-CT and MR imaging, all performed in a radiotherapy treatment mask. CT, MR and CT-MR GTVs were contoured by 5 clinicians (2 radiologists and 3 radiation oncologists). A semi-automated segmentation algorithm was used to contour PET GTVs. Volume and positional analyses were undertaken, accounting for inter-observer variation, using linear mixed effects models and contour comparison metrics respectively. RESULTS: Significant differences in mean GTV volume were found between CT (11.9 cm(3)) and CT-MR (14.1 cm(3)), p < 0.006, CT-MR and PET (9.5 cm(3)), p < 0.0009, and MR (12.7 cm(3)) and PET, p < 0.016. Substantial differences in GTV position were found between all modalities with the exception of CT-MR and MR GTVs. A mean of 64 %, 74 % and 77 % of the PET GTVs were included within the CT, MR and CT-MR GTVs respectively. A mean of 57 % of the MR GTVs were included within the CT GTV; conversely a mean of 63 % of the CT GTVs were included within the MR GTV. CT inter-observer variability was found to be significantly higher in terms of position and/or volume than both MR and CT-MR (p < 0.05). Significant differences in GTV volume were found between GTV volumes delineated by radiologists (9.7 cm(3)) and oncologists (14.6 cm(3)) for all modalities (p = 0.001). CONCLUSIONS: The use of different imaging modalities produced significantly different GTVs, with no single imaging technique encompassing all potential GTV regions. The use of MR reduced inter-observer variability. These data suggest delineation based on multimodality imaging has the potential to improve accuracy of GTV definition. TRIAL REGISTRATION: ISRCTN Registry: ISRCTN34165059 . Registered 2nd February 2015.

Alterations in anatomic and functional imaging parameters with repeated FDG PET-CT and MRI during radiotherapy for head and neck cancer: a pilot study.

BACKGROUND: The use of imaging to implement on-treatment adaptation of radiotherapy is a promising paradigm but current data on imaging changes during radiotherapy is limited. This is a hypothesis-generating pilot study to examine the changes on multi-modality anatomic and functional imaging during (chemo)radiotherapy treatment for head and neck squamous cell carcinoma (HNSCC). METHODS: Eight patients with locally advanced HNSCC underwent imaging including computed tomography (CT), Fluorine-18 fluorodeoxyglucose (FDG) positron emission tomography (PET)-CT and magnetic resonance imaging (MRI) (including diffusion weighted (DW) and dynamic contrast enhanced (DCE)) at baseline and during (chemo)radiotherapy treatment (after fractions 11 and 21). Regions of interest (ROI) were drawn around the primary tumour at baseline and during treatment. Imaging parameters included gross tumour volume (GTV) assessment, SUVmax, mean ADC value and DCE-MRI parameters including Plasma Flow (PF). On treatment changes and correlations between these parameters were analysed using a Wilcoxon rank sum test and Pearson's linear correlation coefficient respectively. A p-value <0.05 was considered statistically significant. RESULTS: Statistically significant reductions in GTV-CT, GTV-MRI and GTV-DW were observed between all imaging timepoints during radiotherapy. Changes in GTV-PET during radiotherapy were heterogeneous and non-significant. Significant changes in SUVmax, mean ADC value, Plasma Flow and Plasma Volume were observed between the baseline and the fraction 11 timepoint, whilst only changes in SUVmax between baseline and the fraction 21 timepoint were statistically significant. Significant correlations were observed between multiple imaging parameters, both anatomical and functional; 20 correlations between baseline to the fraction 11 timepoint; 12 correlations between baseline and the fraction 21 timepoints; and 4 correlations between the fraction 11 and fraction 21 timepoints. CONCLUSIONS: Multi-modality imaging during radiotherapy treatment demonstrates early changes (by fraction 11) in both anatomic and functional imaging parameters. All functional imaging modalities are potentially complementary and should be considered in combination to provide multi-parametric tumour assessment, to guide potential treatment adaptation strategies. TRIAL REGISTRATION: ISRCTN Registry: ISRCTN34165059 . Registered 2nd February 2015.

A Scoping Review of Machine-Learning Derived Radiomic Analysis of CT and PET Imaging to Investigate Atherosclerotic Cardiovascular Disease.

BACKGROUND: Cardiovascular disease affects the carotid arteries, coronary arteries, aorta and the peripheral arteries. Radiomics involves the extraction of quantitative data from imaging features that are imperceptible to the eye. Radiomics analysis in cardiovascular disease has largely focused on CT and MRI modalities. This scoping review aims to summarise the existing literature on radiomic analysis techniques in cardiovascular disease. METHODS: MEDLINE and Embase databases were searched for eligible studies evaluating radiomic techniques in living human subjects derived from CT, MRI or PET imaging investigating atherosclerotic disease. Data on study population, imaging characteristics and radiomics methodology were extracted. RESULTS: Twenty-nine studies consisting of 5753 patients (3752 males) were identified, and 78.7% of patients were from coronary artery studies. Twenty-seven studies employed CT imaging (19 CT carotid angiography and 6 CT coronary angiography (CTCA)), and two studies studied PET/CT. Manual segmentation was most frequently undertaken. Processing techniques included voxel discretisation, voxel resampling and filtration. Various shape, first-order, second-order and higher-order radiomic features were extracted. Logistic regression was most commonly used for machine learning. CONCLUSION: Most published evidence was feasibility/proof of concept work. There was significant heterogeneity in image acquisition, segmentation techniques, processing and analysis between studies. There is a need for the implementation of standardised imaging acquisition protocols, adherence to published reporting guidelines and economic evaluation.

Tumour Size and Overall Survival in a Cohort of Patients with Unifocal Glioblastoma: A Uni- and Multivariable Prognostic Modelling and Resampling Study.

Published models inconsistently associate glioblastoma size with overall survival (OS). This study aimed to investigate the prognostic effect of tumour size in a large cohort of patients diagnosed with GBM and interrogate how sample size and non-linear transformations may impact on the likelihood of finding a prognostic effect. In total, 279 patients with a IDH-wildtype unifocal WHO grade 4 GBM between 2014 and 2020 from a retrospective cohort were included. Uni-/multivariable association between core volume, whole volume (CV and WV), and diameter with OS was assessed with (1) Cox proportional hazard models +/- log transformation and (2) resampling with 1,000,000 repetitions and varying sample size to identify the percentage of models, which showed a significant effect of tumour size. Models adjusted for operation type and a diameter model adjusted for all clinical variables remained significant (p = 0.03). Multivariable resampling increased the significant effects (p < 0.05) of all size variables as sample size increased. Log transformation also had a large effect on the chances of a prognostic effect of WV. For models adjusted for operation type, 19.5% of WV vs. 26.3% log-WV (n = 50) and 69.9% WV and 89.9% log-WV (n = 279) were significant. In this large well-curated cohort, multivariable modelling and resampling suggest tumour volume is prognostic at larger sample sizes and with log transformation for WV.

Functional imaging for radiation treatment planning, response assessment, and adaptive therapy in head and neck cancer.

Patients with squamous cell carcinomas (SCCs) of the head and neck are increasingly treated nonsurgically. Imaging plays a critical role in helping define the targets for radiation therapy, especially intensity-modulated radiation therapy, in which the dose gradients are steep. Anatomic imaging with conventional modalities, particularly computed tomography (CT), has been used in patients with head and neck SCCs, but this approach has limitations. Functional imaging techniques, including positron emission tomography (PET) combined with CT or magnetic resonance (MR) imaging, offer complementary information and can be used noninvasively to assess a range of biomarkers in patients with head and neck SCCs, including hypoxia, cell proliferation and apoptosis, and epidermal growth factor receptor status. These biologic markers can be monitored before, during, and after treatment to improve patient selection for specific therapeutic strategies, guide adaptation of therapy, and potentially facilitate more accurate assessment of disease response. This article discusses the practical aspects of integrating functional imaging into head-and-neck radiation therapy planning and reviews the potential of molecular imaging biomarkers for response assessment and therapy adaptation. The uses of PET tracers for imaging cellular processes such as metabolism, proliferation, hypoxia, and cell membrane synthesis are explored, and applications for MR techniques such as dynamic contrast material-enhanced imaging, diffusion-weighted imaging, blood oxygenation level-dependent imaging, and MR spectroscopy are reviewed. The potential of integrated PET/CT perfusion imaging and hybrid PET/MR imaging also is highlighted. These developments may allow more individualized treatment planning in patients with head and neck SCCs in the emerging era of personalized medicine.

Impact of 18F-fluciclovine PET/CT on plans for androgen deprivation therapy in patients with biochemical recurrence of prostate cancer: data analysis from two prospective clinical trials.

INTRODUCTION: Despite early detection and primary therapy improvements, biochemical recurrence (BCR) of prostate cancer remains common. The advent of highly sensitive molecular imaging has facilitated identification of men with limited metastatic disease burden that might be more optimally treated with metastases-directed therapy than with androgen deprivation therapy (ADT). The LOCATE (NCT02680041) and FALCON (NCT02578940) trials assessed the impact of 18F-fluciclovine PET/CT on the management of patients with BCR after curative-intent primary therapy. We performed a secondary analysis of LOCATE and FALCON data to characterize sites of recurrence and management decisions for BCR patients who had an intended management plan including ADT prior to undergoing 18F-fluciclovine PET/CT. METHODS: Data from 317 LOCATE/FALCON patients who underwent 18F-fluciclovine PET/CT were analyzed and those with a prescan plan for ADT (± another treatment) were selected. 18F-Fluciclovine detection rates were determined at the patient level and for the prostate/prostate bed region, pelvic and extra-pelvic lymph nodes (LN), soft tissues, and bones. The patients' pre- and postscan treatment plans were compared and were stratified by imaging results. RESULTS: A total of 146 patients had a prescan plan for ADT (60 as monotherapy and 86 in combination with another modality). 18F-Fluciclovine detected lesions in 85 of 146 (58%) patients planned for ADT. Detection rates in the prostate/bed, pelvic LN, extra-pelvic LN, soft tissues and bone were 30%, 25%, 13%, 2.1%, and 13%, respectively. Twenty-five (17%) patients had positivity confined to the prostate/bed, 21 (14%) had 18F-fluciclovine-positive pelvic LN (±prostate/bed) but no other involvement and 39 (27%) had involvement outside the prostate/bed and pelvic LN. Postscan, 93 of 146 (64%) patients had a management change, 55 (59%) of which were to abort ADT. Only 25% of the patients originally planned for ADT monotherapy still had an unaltered plan for ADT monotherapy postscan. Patients with a postscan plan for ADT monotherapy had the most disseminated disease. Disease in the prostate/bed only was most common in those whose plan was altered to abort ADT. CONCLUSIONS: 18F-Fluciclovine-PET/CT influenced management plans for the majority of patients with a prescan plan for ADT. Plans were commonly amended to target salvage therapy for lesions identified with 18F-fluciclovine PET/CT, and consequently likely spared/delayed patients the morbidity associated with ADT.

Comparative effectiveness of standard vs. AI-assisted PET/CT reading workflow for pre-treatment lymphoma staging: a multi-institutional reader study evaluation.

Background: Fluorine-18 fluorodeoxyglucose (FDG)-positron emission tomography/computed tomography (PET/CT) is widely used for staging high-grade lymphoma, with the time to evaluate such studies varying depending on the complexity of the case. Integrating artificial intelligence (AI) within the reporting workflow has the potential to improve quality and efficiency. The aims of the present study were to evaluate the influence of an integrated research prototype segmentation tool implemented within diagnostic PET/CT reading software on the speed and quality of reporting with variable levels of experience, and to assess the effect of the AI-assisted workflow on reader confidence and whether this tool influenced reporting behaviour. Methods: Nine blinded reporters (three trainees, three junior consultants and three senior consultants) from three UK centres participated in a two-part reader study. A total of 15 lymphoma staging PET/CT scans were evaluated twice: first, using a standard PET/CT reporting workflow; then, after a 6-week gap, with AI assistance incorporating pre-segmentation of disease sites within the reading software. An even split of PET/CT segmentations with gold standard (GS), false-positive (FP) over-contour or false-negative (FN) under-contour were provided. The read duration was calculated using file logs, while the report quality was independently assessed by two radiologists with >15 years of experience. Confidence in AI assistance and identification of disease was assessed via online questionnaires for each case. Results: There was a significant decrease in time between non-AI and AI-assisted reads (median 15.0 vs. 13.3 min, p < 0.001). Sub-analysis confirmed this was true for both junior (14.5 vs. 12.7 min, p = 0.03) and senior consultants (15.1 vs. 12.2 min, p = 0.03) but not for trainees (18.1 vs. 18.0 min, p = 0.2). There was no significant difference between report quality between reads. AI assistance provided a significant increase in confidence of disease identification (p < 0.001). This held true when splitting the data into FN, GS and FP. In 19/88 cases, participants did not identify either FP (31.8%) or FN (11.4%) segmentations. This was significantly greater for trainees (13/30, 43.3%) than for junior (3/28, 10.7%, p = 0.05) and senior consultants (3/30, 10.0%, p = 0.05). Conclusions: The study findings indicate that an AI-assisted workflow achieves comparable performance to humans, demonstrating a marginal enhancement in reporting speed. Less experienced readers were more influenced by segmentation errors. An AI-assisted PET/CT reading workflow has the potential to increase reporting efficiency without adversely affecting quality, which could reduce costs and report turnaround times. These preliminary findings need to be confirmed in larger studies.

An Automated Method for Artifical Intelligence Assisted Diagnosis of Active Aortitis Using Radiomic Analysis of FDG PET-CT Images.

The aim of this study was to develop and validate an automated pipeline that could assist the diagnosis of active aortitis using radiomic imaging biomarkers derived from [18F]-Fluorodeoxyglucose Positron Emission Tomography-Computed Tomography (FDG PET-CT) images. The aorta was automatically segmented by convolutional neural network (CNN) on FDG PET-CT of aortitis and control patients. The FDG PET-CT dataset was split into training (43 aortitis:21 control), test (12 aortitis:5 control) and validation (24 aortitis:14 control) cohorts. Radiomic features (RF), including SUV metrics, were extracted from the segmented data and harmonized. Three radiomic fingerprints were constructed: A-RFs with high diagnostic utility removing highly correlated RFs; B used principal component analysis (PCA); C-Random Forest intrinsic feature selection. The diagnostic utility was evaluated with accuracy and area under the receiver operating characteristic curve (AUC). Several RFs and Fingerprints had high AUC values (AUC > 0.8), confirmed by balanced accuracy, across training, test and external validation datasets. Good diagnostic performance achieved across several multi-centre datasets suggests that a radiomic pipeline can be generalizable. These findings could be used to build an automated clinical decision tool to facilitate objective and standardized assessment regardless of observer experience.

Dynamic feature learning for COVID-19 segmentation and classification.

Since December 2019, coronavirus SARS-CoV-2 (COVID-19) has rapidly developed into a global epidemic, with millions of patients affected worldwide. As part of the diagnostic pathway, computed tomography (CT) scans are used to help patient management. However, parenchymal imaging findings in COVID-19 are non-specific and can be seen in other diseases. In this work, we propose to first segment lesions from CT images, and further, classify COVID-19 patients from healthy persons and common pneumonia patients. In detail, a novel Dynamic Fusion Segmentation Network (DFSN) that automatically segments infection-related pixels is first proposed. Within this network, low-level features are aggregated to high-level ones to effectively capture context characteristics of infection regions, and high-level features are dynamically fused to model multi-scale semantic information of lesions. Based on DFSN, Dynamic Transfer-learning Classification Network (DTCN) is proposed to distinguish COVID-19 patients. Within DTCN, a pre-trained DFSN is transferred and used as the backbone to extract pixel-level information. Then the pixel-level information is dynamically selected and used to make a diagnosis. In this way, the pre-trained DFSN is utilized through transfer learning, and clinical significance of segmentation results is comprehensively considered. Thus DTCN becomes more sensitive to typical signs of COVID-19. Extensive experiments are conducted to demonstrate effectiveness of the proposed DFSN and DTCN frameworks. The corresponding results indicate that these two models achieve state-of-the-art performance in terms of segmentation and classification.

Impact of 18F-fluciclovine PET/CT on salvage radiotherapy plans for men with recurrence of prostate cancer postradical prostatectomy.

OBJECTIVES: Imaging options to localize biochemical recurrence (BCR) of prostate cancer after radical prostatectomy (RP) are limited, especially at low prostate-specific antigen (PSA) levels. The FALCON study evaluated the impact of 18F-fluciclovine PET/CT on management plans for patients with BCR. Here, we evaluate salvage radiotherapy decisions in patients post-RP. METHODS: We conducted a subgroup analysis of post-RP patients enrolled in FALCON who had a prescan plan for salvage radiotherapy (± androgen-deprivation therapy). Patients' treatment plans post-18F-fluciclovine PET/CT were compared with their prescan plans. Fisher exact test was used to determine the impact of PSA and Gleason sum on positivity and anatomical patterns of uptake. RESULTS: Sixty-five (63%) FALCON patients had undergone RP. Of these, 62 (median PSA, 0.32 ng/mL) had a prescan plan for salvage radiotherapy. Twenty-one (34%) had 18F-fluciclovine-avid lesions. Disease was confined to the prostate bed in 11 patients (52%) and to the pelvis in a further 5 (24%), while 5 (24%) had extrapelvic findings. Trends towards more disseminated disease with increasing PSA or Gleason sum were observed but did not reach statistical significance. Postscan, 25 (40%) patients had a management change; 17 (68%) were changed to the treatment modality (8 to systemic therapy, 8 to active surveillance, 1 other) and 8 (32%) were radiotherapy field modifications. CONCLUSIONS: Incorporating 18F-fluciclovine PET/CT into treatment planning may help identify patients suitable for salvage radiotherapy, help augment planned radiotherapy to better target lesions and support the clinician to optimise patient management.

A Road Map for Designing Phase I Clinical Trials of Radiotherapy-Novel Agent Combinations.

Radiotherapy has proven efficacy in a wide range of cancers. There is growing interest in evaluating radiotherapy-novel agent combinations and a drive to initiate this earlier in the clinical development of the novel agent, where the scientific rationale and preclinical evidence for a radiotherapy combination approach are high. Optimal design, delivery, and interpretation of studies are essential. In particular, the design of phase I studies to determine safety and dosing is critical to an efficient development strategy. There is significant interest in early-phase research among scientific and clinical communities over recent years, at a time when the scrutiny of the trial methodology has significantly increased. To enhance trial design, optimize safety, and promote efficient trial conduct, this position paper reviews the current phase I trial design landscape. Key design characteristics extracted from 37 methodology papers were used to define a road map and a design selection process for phase I radiotherapy-novel agent trials. Design selection is based on single- or dual-therapy dose escalation, dose-limiting toxicity categorization, maximum tolerated dose determination, subgroup evaluation, software availability, and design performance. Fifteen of the 37 designs were identified as being immediately accessible and relevant to radiotherapy-novel agent phase I trials. Applied examples of using the road map are presented. Developing these studies is intensive, highlighting the need for funding and statistical input early in the trial development to ensure appropriate design and implementation from the outset. The application of this road map will improve the design of phase I radiotherapy-novel agent combination trials, enabling a more efficient development pathway.

Discovery of Pre-Treatment FDG PET/CT-Derived Radiomics-Based Models for Predicting Outcome in Diffuse Large B-Cell Lymphoma.

BACKGROUND: Approximately 30% of patients with diffuse large B-cell lymphoma (DLBCL) will have recurrence. The aim of this study was to develop a radiomic based model derived from baseline PET/CT to predict 2-year event free survival (2-EFS). METHODS: Patients with DLBCL treated with R-CHOP chemotherapy undergoing pre-treatment PET/CT between January 2008 and January 2018 were included. The dataset was split into training and internal unseen test sets (ratio 80:20). A logistic regression model using metabolic tumour volume (MTV) and six different machine learning classifiers created from clinical and radiomic features derived from the baseline PET/CT were trained and tuned using four-fold cross validation. The model with the highest mean validation receiver operator characteristic (ROC) curve area under the curve (AUC) was tested on the unseen test set. RESULTS: 229 DLBCL patients met the inclusion criteria with 62 (27%) having 2-EFS events. The training cohort had 183 patients with 46 patients in the unseen test cohort. The model with the highest mean validation AUC combined clinical and radiomic features in a ridge regression model with a mean validation AUC of 0.75 ± 0.06 and a test AUC of 0.73. CONCLUSIONS: Radiomics based models demonstrate promise in predicting outcomes in DLBCL patients.

Salvage Reirradiation Options for Locally Recurrent Prostate Cancer: A Systematic Review.

BACKGROUND: Reirradiation using brachytherapy (BT) and external beam radiation therapy (EBRT) are salvage strategies with locally radiorecurrent prostate cancer. This systematic review describes the oncologic and toxicity outcomes for salvage BT and EBRT [including Stereotactic Body Radiation Therapy (SBRT)]. METHODS: An International Prospective Register of Systematic Reviews (PROSPERO) registered (#211875) study was conducted using Preferred Reporting Items for Systematic reviews and Meta-analyses (PRISMA) guidelines. EMBASE and MEDLINE databases were searched from inception to December 2020. For BT, both low dose rate (LDR) and high dose rate (HDR) BT techniques were included. Two authors independently assessed study quality using the 18-item Modified Delphi technique. RESULTS: A total of 39 eligible studies comprising 1967 patients were included (28 BT and 11 SBRT). In 35 studies (90%), the design was single centre and/or retrospective and no randomised prospective studies were found. Twelve BT studies used LDR only, 11 HDR only, 4 LDR or HDR and 1 pulsed-dose rate only. All EBRT studies used SBRT exclusively, four with Cyberknife alone and 7 using both Cyberknife and conventional linear accelerator treatments. Median (range) modified Delphi quality score was 15 (6-18). Median (range) follow-up was 47.5 months (13-108) (BT) and 25.4 months (21-44) (SBRT). For the LDR-BT studies, the median (range) 2-year and 5-year bRFS rates were 71% (48-89.5) and 52.5% (20-79). For the HDR-BT studies, the median (range) 2-year and 5-year bRFS rates were 74% (63-89) and 51% (45-65). For the SBRT studies, the median (range) 2-year bRFS for the SBRT group was 54.9% (40-80). Mean (range) acute and late grade≥3 GU toxicity rates for LDR-BT/HDR-BT/SBRT were 7.4%(0-14)/2%(0-14)/2.7%(0-8.7) and 13.6%(0-30)/7.9%(0-21.3%)/2.7%(0-8%). Mean (range) acute and late grade≥3 GI toxicity rates for LDR-BT/HDR-BT/SBRT were 6.5%(0-19)/0%/0.5%(0-4%) and 6.4%(0-20)/0.1%(0-0.9)/0.2%(0-1.5). One third of studies included Patient Reported Outcome Measures (PROMs). CONCLUSIONS: Salvage reirradiation of radiorecurrent prostate cancer using HDR-BT or SBRT provides similar biochemical control and acceptable late toxicity. Salvage LDR-BT is associated with higher late GU/GI toxicity. Challenges exist in comparing BT and SBRT from inconsistencies in reporting with missing data, and prospective randomised trials are needed.