Challenges and opportunities in the development and clinical implementation of artificial intelligence based synthetic computed tomography for magnetic resonance only radiotherapy.

Synthetic computed tomography (sCT) generated from magnetic resonance imaging (MRI) can serve as a substitute for planning CT in radiation therapy (RT), thereby removing registration uncertainties associated with multi-modality imaging pairing, reducing costs and patient radiation exposure. CE/FDA-approved sCT solutions are nowadays available for pelvis, brain, and head and neck, while more complex deep learning (DL) algorithms are under investigation for other anatomic sites. The main challenge in achieving a widespread clinical implementation of sCT lies in the absence of consensus on sCT commissioning and quality assurance (QA), resulting in variation of sCT approaches across different hospitals. To address this issue, a group of experts gathered at the ESTRO Physics Workshop 2022 to discuss the integration of sCT solutions into clinics and report the process and its outcomes. This position paper focuses on aspects of sCT development and commissioning, outlining key elements crucial for the safe implementation of an MRI-only RT workflow.

Thyroglobulin measurement is the most powerful outcome predictor in differentiated thyroid cancer: a decision tree analysis in a European multicenter series.

OBJECTIVES: An accurate prognostic assessment is pivotal to adequately inform and individualize follow-up and management of patients with differentiated thyroid cancer (DTC). We aimed to develop a predictive model for recurrent disease in DTC patients treated by surgery and 131I by adopting a decision tree model. METHODS: Age, sex, histology, T stage, N stage, risk classes, remnant estimation, thyroid-stimulating hormone (TSH), thyroglobulin (Tg), administered 131I activities and post-therapy whole body scintigraphy (PT-WBS) were identified as potential predictors and put into regression algorithm (conditional inference tree, c-tree) to develop a risk stratification model for predicting persistent/recurrent disease over time. RESULTS: The PT-WBS pattern identified a partition of the population into two subgroups (PT-WBS positive or negative for distant metastases). Patients with distant metastases exhibited lower disease-free survival (either structural, DFS-SD, and biochemical, DFS-BD, disease) compared to those without metastases. Meanwhile, the latter were further stratified into three risk subgroups based on their Tg values. Notably, Tg values >63.1 ng/mL predicted a shorter survival time, with increased DFS-SD for Tg values <63.1 and <8.9 ng/mL, respectively. A comparable model was generated for biochemical disease (BD), albeit different DFS were predicted by slightly different Tg cutoff values (41.2 and 8.8 ng/mL) compared to DFS-SD. CONCLUSIONS: We developed a simple, accurate and reproducible decision tree model able to provide reliable information on the probability of structurally and/or biochemically persistent/relapsed DTC after a TTA. In turn, the provided information is highly relevant to refine the initial risk stratification, identify patients at higher risk of reduced structural and biochemical DFS, and modulate additional therapies and the relative follow-up.

Postoperative thyroglobulin as a yard-stick for radioiodine therapy: decision tree analysis in a European multicenter series of 1317 patients with differentiated thyroid cancer.

PURPOSE: An accurate postoperative assessment is pivotal to inform postoperative 131I treatment in patients with differentiated thyroid cancer (DTC). We developed a predictive model for post-treatment whole-body scintigraphy (PT-WBS) results (as a proxy for persistent disease) by adopting a decision tree model. METHODS: Age, sex, histology, T stage, N stage, risk classes, remnant estimation, TSH, and Tg were identified as potential predictors and were put into regression algorithm (conditional inference tree, ctree) to develop a risk stratification model for predicting the presence of metastases in PT-WBS. RESULTS: The lymph node (N) stage identified a partition of the population into two subgroups (N-positive vs N-negative). Among N-positive patients, a Tg value > 23.3 ng/mL conferred a 83% probability to have metastatic disease compared to those with lower Tg values. Additionally, N-negative patients were further substratified in three subgroups with different risk rates according to their Tg values. The model remained stable and reproducible in the iterative process of cross validation. CONCLUSIONS: We developed a simple and robust decision tree model able to provide reliable informations on the probability of persistent/metastatic DTC after surgery. These information may guide post-surgery 131I administration and select patients requiring curative rather than adjuvant 131I therapy schedules.

CT evaluation of lung infiltrates in the two months preceding the Coronavirus disease 19 pandemic in Canton Ticino (Switzerland): were there suspicious cases before the official first case?

PURPOSE: The main objective of this study was to assess the presence of pulmonary infiltrates with computed tomography (CT) appearance compatible with infection by coronavirus disease 2019 (COVID-19), in Canton Ticino in the 2 months preceding the first official case. Secondary aims were to compare the classification of infiltrates in the same time frame in 2020 and 2019; to compare the number of chest CT scans in the same period; to search for pathological confirmation of the virus. MATERIALS AND METHODS: Chest CT scans performed between January 1 and February 24 in 2019 and 2020 were collected and classified by COVID-19 Reporting and Data System (CO-RADS). Pathological presence of the virus was searched for when appropriate material was available. RESULTS: The final cohort included 881 patients. Among the CO-RADS 3 and 4 categories, 30 patients had pneumonitis of unknown etiology. Pathological specimens were available in six patients but they were negative for COVID-19. CONCLUSION: Before the first official case of COVID-19 infection, in Canton Ticino there were about 30 cases of pneumonitis of uncertain origin, with CT appearance compatible with infection by COVID-19, but with no confirmation of the disease. The number of chest CT scans in the first two months of 2020 was > 12% compared to 2019.

Integration of Baseline Metabolic Parameters and Mutational Profiles Predicts Long-Term Response to First-Line Therapy in DLBCL Patients: A Post Hoc Analysis of the SAKK38/07 Study.

Accurate estimation of the progression risk after first-line therapy represents an unmet clinical need in diffuse large B-cell lymphoma (DLBCL). Baseline (18)F-fluorodeoxyglucose positron emission tomography/computed tomography (PET/CT) parameters, together with genetic analysis of lymphoma cells, could refine the prediction of treatment failure. We evaluated the combined impact of mutation profiling and baseline PET/CT functional parameters on the outcome of DLBCL patients treated with the R-CHOP14 regimen in the SAKK38/07 clinical trial (NCT00544219). The concomitant presence of mutated SOCS1 with wild-type CREBBP and EP300 defined a group of patients with a favorable prognosis and 2-year progression-free survival (PFS) of 100%. Using an unsupervised recursive partitioning approach, we generated a classification-tree algorithm that predicts treatment outcomes. Patients with elevated metabolic tumor volume (MTV) and high metabolic heterogeneity (MH) (15%) had the highest risk of relapse. Patients with low MTV and favorable mutational profile (9%) had the lowest risk, while the remaining patients constituted the intermediate-risk group (76%). The resulting model stratified patients among three groups with 2-year PFS of 100%, 82%, and 42%, respectively (p < 0.001).

Radiomics analysis improves 18FDG PET/CT-based risk stratification of cytologically indeterminate thyroid nodules.

PURPOSE: As ~25% of cytologically indeterminate thyroid nodules harbour malignancy, diagnostic lobectomy is still performed in many cases. 18FDG PET/CT rules out malignancy in visually negative nodules; however, none of the currently available interpretation criteria differentiates malignant from benign 18FDG-avid nodules. We evaluated the ability of PET metrics and radiomics features (RFs) to predict final diagnosis of 18FDG-avid cytologically indeterminate thyroid nodules. METHODS: Seventy-eight patients were retrospectively included. After volumetric segmentation of each thyroid lesion, 4 PET metrics and 107 RFs were extracted. A logistic regression was performed including thyroid stimulating hormone, PET metrics, and RFs to assess their predictive performance. A linear combination of the resulting parameters generated a radiomics score (RS) that was matched with cytology classes (Bethesda III and IV) and compared with final diagnosis. RESULTS: Two RFs (shape_Sphericity and glcm_Autocorrelation) differentiated malignant from benign lesions. A predictive model integrating RS and cytology classes effectively stratified the risk of malignancy. The prevalence of thyroid cancer increased from 5 to 37% and 79% in accordance with the number (score 0, 1 or 2, respectively) of positive biomarkers. CONCLUSIONS: Our multiparametric model may be useful for reducing the number of diagnostic lobectomies with advantages in terms of costs and quality of life for patients.

Generation and validation of a PET radiomics model that predicts survival in diffuse large B cell lymphoma treated with R-CHOP14: A SAKK 38/07 trial post-hoc analysis.

Functional parameters from positron emission tomography (PET) seem promising biomarkers in various lymphoma subtypes. This study investigated the prognostic value of PET radiomics in diffuse large B-cell lymphoma (DLBCL) patients treated with R-CHOP given either every 14 (testing set) or 21 days (validation set). Using the PyRadiomics Python package, 107 radiomics features were extracted from baseline PET scans of 133 patients enrolled in the Swiss Group for Clinical Cancer Research 38/07 prospective clinical trial (SAKK 38/07) [ClinicalTrial.gov identifier: NCT00544219]. The international prognostic indices, the main clinical parameters and standard PET metrics, together with 52 radiomics uncorrelated features (selected using the Spearman correlation test) were included in a least absolute shrinkage and selection operator (LASSO) Cox regression to assess their impact on progression-free (PFS), cause-specific (CSS), and overall survival (OS). A linear combination of the resulting parameters generated a prognostic radiomics score (RS) whose area under the curve (AUC) was calculated by receiver operating characteristic analysis. The RS efficacy was validated in an independent cohort of 107 DLBCL patients. LASSO Cox regression identified four radiomics features predicting PFS in SAKK 38/07. The derived RS showed a significant capability to foresee PFS in both testing (AUC, 0.709; p < 0.001) and validation (AUC, 0.706; p < 0.001) sets. RS was significantly associated also with CSS and OS in testing (CSS: AUC, 0.721; p < 0.001; OS: AUC, 0.740; p < 0.001) and validation (CSS: AUC, 0.763; p < 0.0001; OS: AUC, 0.703; p = 0.004) sets. The RS allowed risk classification of patients with significantly different PFS, CSS, and OS in both cohorts showing better predictive accuracy respect to clinical international indices. PET-derived radiomics may improve the prediction of outcome in DLBCL patients.

A single phantom, a single statistical method for low-contrast detectability assessment.

PURPOSE: The assessment of low-contrast-details is a part of the quality control (QC) program in digital radiology. It generally consists of evaluating the threshold contrast (Cth) detectability details for different-sized inserts, appropriately located in dedicated QC test tools. This work aims to propose a simplified method, based on a statistical model approach for threshold contrast estimation, suitable for different modalities in digital radiology. METHODS: A home-madelow-contrast phantom, made of a central aluminium insert with a step-wedge, was assembled and tested. The reliability and robustness of the method were investigated for Mammography, Digital Radiography, Fluoroscopy and Angiography. Imageswere analysed using our dedicated software developed on Matlab®. TheCth is expressed in the same unit (mmAl) for all studied modalities. RESULTS: This method allows the collection of Cthinformation from different modalities and equipment by different vendors, and it could be used to define typical values. Results are summarized in detail. For 0.5 diameter detail, Cthresults are in the range of: 0.018-0.023 mmAl for 2D mammography and 0.26-0.34 mmAl DR images. For angiographic images, for 2.5 mm diameter detail, the Cths median values are 0.55, 0.4, 0.06, 0.12 mmAl for low dose fluoroscopy, coronary fluorography, cerebral and abdominal DSA, respectively. CONCLUSIONS: The statistical method proposed in this study gives a simple approach for Low-Contrast-Details assessment, and the typical values proposed can be implemented in a QA program for digital radiology modalities.

Artificial intelligence applications in medical imaging: A review of the medical physics research in Italy.

PURPOSE: To perform a systematic review on the research on the application of artificial intelligence (AI) to imaging published in Italy and identify its fields of application, methods and results. MATERIALS AND METHODS: A Pubmed search was conducted using terms Artificial Intelligence, Machine Learning, Deep learning, imaging, and Italy as affiliation, excluding reviews and papers outside time interval 2015-2020. In a second phase, participants of the working group AI4MP on Artificial Intelligence of the Italian Association of Physics in Medicine (AIFM) searched for papers on AI in imaging. RESULTS: The Pubmed search produced 794 results. 168 studies were selected, of which 122 were from Pubmed search and 46 from the working group. The most used imaging modality was MRI (44%) followed by CT(12%) ad radiography/mammography (11%). The most common clinical indication were neurological diseases (29%) and diagnosis of cancer (25%). Classification was the most common task for AI (57%) followed by segmentation (16%). 65% of studies used machine learning and 35% used deep learning. We observed a rapid increase of research in Italy on artificial intelligence in the last 5 years, peaking at 155% from 2018 to 2019. CONCLUSIONS: We are witnessing an unprecedented interest in AI applied to imaging in Italy, in a diversity of fields and imaging techniques. Further initiatives are needed to build common frameworks and databases, collaborations among different types of institutions, and guidelines for research on AI.

Determining an energy threshold for optimal volume reduction of benign thyroid nodules treated by radiofrequency ablation.

OBJECTIVES: Radiofrequency ablation (RFA) is effective in reducing the volume of benign thyroid nodules. However, what parameters can influence the response to RFA is still unclear. The present study aimed to (1) investigate which ultrasound and technical parameters are potential determinants of the volumetric reduction; (2) develop a dose-response model, and (3) analyze the effects of RFA on ultrasound features. METHODS: In this retrospective study, three institutions treated patients with benign thyroid nodules according to the same protocol. The technical parameters were power and energy. The 1-year volume reduction ratio (VRR) was the reference standard of the response. The correlations of different parameters with VRR were analyzed and the association between several parameters and a VRR above 50% studied by uni- and multivariate analyses. The probit regression estimated the probability to achieve an effective response. RESULTS: One hundred fifteen patients were enrolled. The median power was 50 W and median total delivered energy 27,531 J. At 1-year follow-up, the median VRR was 64.0% and 87 (75.7%) nodules showed a VRR above 50%. Among all parameters, only baseline volume, total energy, and energy per volume were independently associated to a VRR > 50% (p = 0.001, p = 0.0178, p < 0.001 respectively). The probit regression analysis demonstrated that delivering 756 J/ml and 2670 J/ml gave a probability of VRR > 50% in 50% and 99% of patients, respectively. CONCLUSIONS: Considering the baseline nodular volume and delivering the adequate energy per volume allow optimizing technical and clinical success. KEY POINTS: • The effectiveness of radiofrequency ablation in treating benign thyroid nodules is negatively correlated to the volume of the nodule and positively correlated to the energy delivered per volume. • When planning the treatment, the total energy to deliver can be calculated by using a simple formula: nodular volume × 2670 J.

Radiomics Analysis of [18F]-Fluorodeoxyglucose-Avid Thyroid Incidentalomas Improves Risk Stratification and Selection for Clinical Assessment.

Background: [18F]-Fluorodeoxyglucose (FDG)-avid thyroid lesions incidentally detected on positron emission tomography/computed tomography (PET/CT) scans represent a tumor lesion in about 30% of cases. The present study evaluated the ability of PET metrics and radiomics features to predict final diagnosis of [18F]FDG thyroid incidentalomas (TIs). Methods: A total of 104 patients with 107 TIs were retrospectively studied; 30 nodules (28%) were diagnosed as malignant. After volumetric segmentation of each thyroid lesion, metabolic tumor volume (MTV), total lesion glycolysis (TLG), standardized uptake values (SUVs), and metabolic heterogeneity were estimated, and 107 radiomics features were extracted following a standard protocol. Results: MTV, TLG, SUVmax, SUVmean, and SUVpeak among functional PET parameters, and gray-level co-occurrence matrix (GLCM)_InverseDifferenceMoment, shape_Sphericity, GLCM_SumSquares, firstorder_Maximum2DDiameterSlice, firstorder_Energy, and GLCM_Contrast among nonredundant radiomics features, showed significantly different values between malignant and benign TIs (Mann-Whitney U-test, p < 0.01 for all). Univariate logistic regression revealed that these parameters demonstrated good ability to predict final diagnosis of TIs (p < 0.02 for all). Shape_Sphericity was the best predictor classifying 82% of TIs correctly (p < 0.0001). Only TLG, SUVmax, and shape_Sphericity retained significance (p < 0.0001) by multivariate analysis. Malignant lesion prevalence increased from 7% to 100% in accordance with the number (score, 0-3) of the three positive parameters present (χ2 trend, p < 0.0001). A score of 0 excludes malignant TIs with a negative predictive value of 93%, while a score of 3 predicted malignancy with a positive predictive value of 100%. Conclusions: PET metrics and radiomics analysis can improve identification of [18F]FDG-avid TIs at high risk of malignancy. A model based on TLG, SUVmax, and shape_Sphericity may allow prediction of a final diagnosis, providing useful information for the management of TIs.

Baseline PET features to predict prognosis in primary mediastinal B cell lymphoma: a comparative analysis of different methods for measuring baseline metabolic tumour volume.

PURPOSE: This study assessed the performance of four different methods for the estimation of metabolic tumour volume (MTV) in primary mediastinal B cell lymphoma (PMBCL). METHOD: MTV was estimated using either a region growing automatic software program (RG) or a fixed threshold (FT) segmentation algorithm with the three most common cut-offs proposed in the literature (i.e., 25% and 41% of the SUVmax and SUV value ≥2.5). We compared these four methods using phantoms that simulated different set-ups of the main imaging characteristics of PMBCL (volume, shape, 18-FDG uptake and intra-lesion distribution) and assessed their performance in 103 PMBCL patients enrolled in the International Extranodal Lymphoma Study Group-26 (IELSG-26) study. RESULTS: There was good correlation between MTV values estimated in vitro and in vivo using the different methods. The 25% FT cut-off (FT25%) provided the most accurate MTV evaluation in the phantoms. The cut-off at SUV 2.5 (FT2.5) resulted in MTV overestimation that particularly increased with high SUV values. The 41% cut-off (FT41%) showed MTV underestimation that was more evident when there were high levels of heterogeneity in tracer distribution. Shape of the lesion did not affect MTV computation. The RG algorithm provided a systematic slight MTV underestimation without significant changes due to lesion characteristics. We observed analogous trends for the MTV estimation in patients, with very different derived thresholds for the four methods. Optimal cut-offs for predicting progression-free survival (PFS) ranged from 213 to 831 ml. All methods predicted PFS with similar negative predictive values (94-95%) but different positive predictive values (23-45%). CONCLUSIONS: The different methods result in significantly different MTV cut-off values. All allow risk stratification in PMBCL, but FT25% showed the best capacity to predict disease progression in the patient cohort and provided the best accuracy in the phantom model.

Metabolic heterogeneity on baseline 18FDG-PET/CT scan is a predictor of outcome in primary mediastinal B-cell lymphoma.

An important unmet need in the management of primary mediastinal B-cell lymphoma (PMBCL) is to identify the patients for whom first-line therapy will fail to intervene before the lymphoma becomes refractory. High heterogeneity of intratumoral 18F-fluorodeoxyglucose (18FDG) uptake distribution on positron emission tomography/computed tomography (PET/CT) scans has been suggested as a possible marker of chemoresistance in solid tumors. In the present study, we investigated the prognostic value of metabolic heterogeneity (MH) in 103 patients with PMBCL prospectively enrolled in the International Extranodal Lymphoma Study Group (IELSG) 26 study, aimed at clarifying the role of PET in this lymphoma subtype. MH was estimated using the area under curve of cumulative standardized uptake value-volume histogram (AUC-CSH) method. Progression-free survival at 5 years was 94% vs 73% in low- and high-MH groups, respectively (P = .0001). In a Cox model of progression-free survival including dichotomized MH, metabolic tumor volume, total lesion glycolysis (TLG), international prognostic index, and tumor bulk (mediastinal mass > 10 cm), as well as age as a continuous variable, only TLG (P < .001) and MH (P < .001) retained statistical significance. Using these 2 features to construct a simple prognostic model resulted in early and accurate (positive predictive value, 89%; negative predictive value, ≥90%) identification of patients at high risk for progression at a point that would allow the use of risk-adapted treatments. This may provide an important opportunity for the design of future trials aimed at helping the minority of patients who harbor chemorefractory PMBCL. The study is registered at ClinicalTrials.gov as NCT00944567.