Glioblastoma and radiotherapy: A multicenter AI study for Survival Predictions from MRI (GRASP study).

BACKGROUND: The aim was to predict survival of glioblastoma at 8 months after radiotherapy (a period allowing for completing a typical course of adjuvant temozolomide), by applying deep learning to the first brain MRI after radiotherapy completion. METHODS: Retrospective and prospective data were collected from 206 consecutive glioblastoma, isocitrate dehydrogenase -wildtype patients diagnosed between March 2014 and February 2022 across 11 UK centers. Models were trained on 158 retrospective patients from 3 centers. Holdout test sets were retrospective (n = 19; internal validation), and prospective (n = 29; external validation from 8 distinct centers). Neural network branches for T2-weighted and contrast-enhanced T1-weighted inputs were concatenated to predict survival. A nonimaging branch (demographics/MGMT/treatment data) was also combined with the imaging model. We investigated the influence of individual MR sequences; nonimaging features; and weighted dense blocks pretrained for abnormality detection. RESULTS: The imaging model outperformed the nonimaging model in all test sets (area under the receiver-operating characteristic curve, AUC P = .038) and performed similarly to a combined imaging/nonimaging model (P > .05). Imaging, nonimaging, and combined models applied to amalgamated test sets gave AUCs of 0.93, 0.79, and 0.91. Initializing the imaging model with pretrained weights from 10 000s of brain MRIs improved performance considerably (amalgamated test sets without pretraining 0.64; P = .003). CONCLUSIONS: A deep learning model using MRI images after radiotherapy reliably and accurately determined survival of glioblastoma. The model serves as a prognostic biomarker identifying patients who will not survive beyond a typical course of adjuvant temozolomide, thereby stratifying patients into those who might require early second-line or clinical trial treatment.

Pilot study of lung cancer screening for survivors of Hodgkin lymphoma.

Hodgkin lymphoma (HL) treatment increases the risk of lung cancer. Most HL survivors are not eligible for lung cancer screening (LCS) programmes developed for the general population, and the utility of these programmes has not been tested in HL survivors. We ran a LCS pilot in HL survivors to describe screening uptake, participant characteristics, impact of a decision aid and screen findings. HL survivors treated ≥5 years ago with mustine/procarbazine and/or thoracic radiation, were identified from a follow-up database and invited to participate. Participants underwent a low-dose CT (LDCT) reported using protocols validated for the general population. Two hundred and eighteen individuals were invited, 123 were eligible, 102 were screened (58% response rate): 58% female, median age 52 years, median 22 years since HL treatment. 91.4% were deemed to have made an informed decision; participation was not influenced by age, gender, years since treatment or deprivation. Only 3/35 ever-smokers met criteria for LCS through the programme aimed at the general population. Baseline LDCT results were: 90 (88.2%) negative, 10 (9.8%) indeterminate, 2 (2.0%) positive. Two 3-month surveillance scans were positive. Of 4 positive scans, 2 patients were diagnosed with small-cell lung cancer; 1 underwent curative surgery. Coronary artery calcification was detected in 36.3%, and clinically significant incidental findings in 2.9%. LDCT protocols validated in ever-smokers can detect asymptomatic early-stage lung cancers in HL survivors. This finding, together with screening uptake and low false positive rates, supports further research to implement LCS for HL survivors.

Overcoming challenges of translating deep-learning models for glioblastoma: the ZGBM consortium.

OBJECTIVE: To report imaging protocol and scheduling variance in routine care of glioblastoma patients in order to demonstrate challenges of integrating deep-learning models in glioblastoma care pathways. Additionally, to understand the most common imaging studies and image contrasts to inform the development of potentially robust deep-learning models. METHODS: MR imaging data were analysed from a random sample of five patients from the prospective cohort across five participating sites of the ZGBM consortium. Reported clinical and treatment data alongside DICOM header information were analysed to understand treatment pathway imaging schedules. RESULTS: All sites perform all structural imaging at every stage in the pathway except for the presurgical study, where in some sites only contrast-enhanced T1-weighted imaging is performed. Diffusion MRI is the most common non-structural imaging type, performed at every site. CONCLUSION: The imaging protocol and scheduling varies across the UK, making it challenging to develop machine-learning models that could perform robustly at other centres. Structural imaging is performed most consistently across all centres. ADVANCES IN KNOWLEDGE: Successful translation of deep-learning models will likely be based on structural post-treatment imaging unless there is significant effort made to standardise non-structural or peri-operative imaging protocols and schedules.

Incidental pulmonary emboli are associated with a very high probability of progressive malignant disease on staging CT scans.

PURPOSE: The link between malignancy and venous thromboembolism (VTE) is well established; the risk has been shown to be higher in certain tumour types and in patients receiving platinum-based chemotherapy regimens. 'Active' cancer is often quoted as being a particular risk factor for VTE, but the definition varies widely and the supporting evidence is weak. This study was conceived to establish whether incidental pulmonary emboli are associated with an increased likelihood of progressive disease on a staging computed tomography (CT) scan. MATERIALS AND METHODS: All CT scans containing incidental PEs over the course of a year in a large specialist cancer hospital were examined (n = 306). An age-matched control group was assembled. Multivariate logistic regression analysis was performed to establish the relationship of several independent variables, including progressive disease, with the dependent variable of the presence or absence of incidental pulmonary emboli. RESULTS: Progressive disease was present in 144 of 306 (47.1%) of the patients in the PE group but only 63 of 306 (20.6%) of the control group patients. Progressive disease on the staging CT scan was found to have an odds ratio for incidental PE of 3.46, with a 95% CI of 2.38-5.01. Receiving platinum-based chemotherapy resulted in an odds ratio of 3.89 (2.41-6.28) for PE. Receiving non-platinum-based systemic anti-cancer therapy (SACT) resulted in an odds ratio for PE of 1.71 (1.16-2.50). CONCLUSION: The detection of an incidental PE on a staging CT scan is associated with a very high risk of progressive malignant disease.

Prospective intra-individual comparison of standard dose versus reduced-dose thoracic CT using hybrid and pure iterative reconstruction in a follow-up cohort of pulmonary nodules-Effect of detectability of pulmonary nodules with lowering dose based on nodule size, type and body mass index.

OBJECTIVES: To determine the diagnostic accuracy of lung nodule detection in thoracic CT using 2 reduced dose protocols comparing 3 available CT reconstruction algorithms (filtered back projection-FBP, adaptive statistical reconstruction-ASIR and model-based iterative reconstruction-MBIR) in a western population. MATERIALS AND METHODS: A prospective single-center study recruited 98 patients with written consent. Standard dose (STD) thoracic CT followed by 2 reduced-dose protocols using automatic tube current modulation (RD1) and fixed tube current (RD2) were performed and reconstructed with FBP, ASIR and MBIR with subsequent diagnostic accuracy analysis for nodule detection. RESULTS: 108 solid nodules, 47 subsolid nodules and 89 purely calcified nodules were analyzed. RD1 was superior to RD2 for assessment of solid nodules ≤4mm, and subsolid nodules ≤5mm (p<0.05). Deterioration of RD2 is correlated to patient's body mass index and least affected by MBIR. For solid nodules ≤4mm, MBIR area under curve (AUC) for RD1 was 0.935/0.913 and AUC for RD2 was 0.739/0.739, for rater 1/rater2 respectively. For subsolid nodules ≤5mm, MBIR AUC for RD1 was 0.971/0.986 and AUC for RD2 was 0.914/0.914, for rater 1/rater2 respectively. For calcified nodules excellent detection accuracy was maintained regardless of reconstruction algorithms with AUC >0.97 for both readers across all dose and reconstruction algorithms. CONCLUSIONS: Diagnostic performance of lung nodule is affected by nodule size, protocol, reconstruction algorithm and patient's body habitus. The protocol in this study showed that RD1 was superior to RD2 for assessment of solid nodules ≤4mm, and subsolid nodules ≤5mm and deterioration of RD2 is related to patient's body mass index.

Intra-patient comparison of reduced-dose model-based iterative reconstruction with standard-dose adaptive statistical iterative reconstruction in the CT diagnosis and follow-up of urolithiasis.

OBJECTIVES: To evaluate the accuracy of reduced-dose CT scans reconstructed using a new generation of model-based iterative reconstruction (MBIR) in the imaging of urinary tract stone disease, compared with a standard-dose CT using 30% adaptive statistical iterative reconstruction. METHODS: This single-institution prospective study recruited 125 patients presenting either with acute renal colic or for follow-up of known urinary tract stones. They underwent two immediately consecutive scans, one at standard dose settings and one at the lowest dose (highest noise index) the scanner would allow. The reduced-dose scans were reconstructed using both ASIR 30% and MBIR algorithms and reviewed independently by two radiologists. Objective and subjective image quality measures as well as diagnostic data were obtained. RESULTS: The reduced-dose MBIR scan was 100% concordant with the reference standard for the assessment of ureteric stones. It was extremely accurate at identifying calculi of 3 mm and above. The algorithm allowed a dose reduction of 58% without any loss of scan quality. CONCLUSIONS: A reduced-dose CT scan using MBIR is accurate in acute imaging for renal colic symptoms and for urolithiasis follow-up and allows a significant reduction in dose. KEY POINTS: • MBIR allows reduced CT dose with similar diagnostic accuracy • MBIR outperforms ASIR when used for the reconstruction of reduced-dose scans • MBIR can be used to accurately assess stones 3 mm and above.

The clinical use of contrast-enhanced ultrasound in the kidney.

Traditional B-Mode and Doppler sonography have been the stalwart of renal tract imaging for many years, and indeed, are in daily use in most centres as the modality of choice for the initial assessment of renal pathology. However, traditional ultrasound scanning can be limited in its ability to accurately characterise renal pathology, and can be inaccurate at determining benign from malignant lesions. Contrast-enhanced ultrasound conveys many benefits, being safe (especially in patients with renal dysfunction), does not require the use of ionising radiation, is quick and relatively cheap and can help to establish whether a focal renal lesion is sinister. Furthermore, it is our experience that contrast-enhanced ultrasound is not a difficult technique to master for the experienced ultrasound practitioner. In this article, we discuss the technique, interpretation and value of contrast-enhanced ultrasound in renal imaging, and describe how we use it in our practice.

Lithium rescues toxicity of aggregate-prone proteins in Drosophila by perturbing Wnt pathway.

We have previously shown that lithium can protect against the polyglutamine toxicity of the Huntington's disease mutation in cell models. Here, we demonstrate for the first time in vivo that lithium can protect against the toxicity caused by aggregate-prone proteins with either polyglutamine or polyalanine expansions in Drosophila. We also show that these protective effects can be partly accounted for by lithium acting through the Wnt/Wg pathway, as a GSK3beta-specific inhibitor and overexpression of dTCF also mediate protective effects. Our data suggest that lithium deserves serious consideration for further studies as a therapeutic for polyglutamine diseases, particularly as it is an established drug that has been used for several decades for chronic treatment of affective disorders.