Radiomics relies on the extraction of a wide variety of quantitative image-based features to provide decision support. Magnetic resonance imaging (MRI) contributes to the personalization of patient ...care but suffers from being highly dependent on acquisition and reconstruction parameters. Today, there are no guidelines regarding the optimal pre-processing of MR images in the context of radiomics, which is crucial for the generalization of published image-based signatures. This study aims to assess the impact of three different intensity normalization methods (Nyul, WhiteStripe, Z-Score) typically used in MRI together with two methods for intensity discretization (fixed bin size and fixed bin number). The impact of these methods was evaluated on first- and second-order radiomics features extracted from brain MRI, establishing a unified methodology for future radiomics studies. Two independent MRI datasets were used. The first one (DATASET1) included 20 institutional patients with WHO grade II and III gliomas who underwent post-contrast 3D axial T1-weighted (T1w-gd) and axial T2-weighted fluid attenuation inversion recovery (T2w-flair) sequences on two different MR devices (1.5 T and 3.0 T) with a 1-month delay. Jensen-Shannon divergence was used to compare pairs of intensity histograms before and after normalization. The stability of first-order and second-order features across the two acquisitions was analysed using the concordance correlation coefficient and the intra-class correlation coefficient. The second dataset (DATASET2) was extracted from the public TCIA database and included 108 patients with WHO grade II and III gliomas and 135 patients with WHO grade IV glioblastomas. The impact of normalization and discretization methods was evaluated based on a tumour grade classification task (balanced accuracy measurement) using five well-established machine learning algorithms. Intensity normalization highly improved the robustness of first-order features and the performances of subsequent classification models. For the T1w-gd sequence, the mean balanced accuracy for tumour grade classification was increased from 0.67 (95% CI 0.61-0.73) to 0.82 (95% CI 0.79-0.84, P = .006), 0.79 (95% CI 0.76-0.82, P = .021) and 0.82 (95% CI 0.80-0.85, P = .005), respectively, using the Nyul, WhiteStripe and Z-Score normalization methods compared to no normalization. The relative discretization makes unnecessary the use of intensity normalization for the second-order radiomics features. Even if the bin number for the discretization had a small impact on classification performances, a good compromise was obtained using the 32 bins considering both T1w-gd and T2w-flair sequences. No significant improvements in classification performances were observed using feature selection. A standardized pre-processing pipeline is proposed for the use of radiomics in MRI of brain tumours. For models based on first- and second-order features, we recommend normalizing images with the Z-Score method and adopting an absolute discretization approach. For second-order feature-based signatures, relative discretization can be used without prior normalization. In both cases, 32 bins for discretization are recommended. This study may pave the way for the multicentric development and validation of MR-based radiomics biomarkers.
Abstract
The use of multicentric data is becoming essential for developing generalizable radiomic signatures. In particular, Magnetic Resonance Imaging (MRI) data used in brain oncology are often ...heterogeneous in terms of scanners and acquisitions, which significantly impact quantitative radiomic features. Various methods have been proposed to decrease dependency, including methods acting directly on MR images, i.e., based on the application of several preprocessing steps before feature extraction or the ComBat method, which harmonizes radiomic features themselves. The ComBat method used for radiomics may be misleading and presents some limitations, such as the need to know the labels associated with the “batch effect”. In addition, a statistically representative sample is required and the applicability of a signature whose batch label is not present in the train set is not possible. This work aimed to compare a priori and a posteriori radiomic harmonization methods and propose a code adaptation to be machine learning compatible. Furthermore, we have developed AutoComBat, which aims to automatically determine the batch labels, using either MRI metadata or quality metrics as inputs of the proposed constrained clustering. A heterogeneous dataset consisting of high and low-grade gliomas coming from eight different centers was considered. The different methods were compared based on their ability to decrease relative standard deviation of radiomic features extracted from white matter and on their performance on a classification task using different machine learning models. ComBat and AutoComBat using image-derived quality metrics as inputs for batch assignment and preprocessing methods presented promising results on white matter harmonization, but with no clear consensus for all MR images. Preprocessing showed the best results on the T1w-gd images for the grading task. For T2w-flair, AutoComBat, using either metadata plus quality metrics or metadata alone as inputs, performs better than the conventional ComBat, highlighting its potential for data harmonization. Our results are MRI weighting, feature class and task dependent and require further investigations on other datasets.
Abstract
In glioblastoma, the response to treatment assessment is essentially based on the 2D tumor size evolution but remains disputable. Volumetric approaches were evaluated for a more accurate ...estimation of tumor size. This study included 57 patients and compared two volume measurement methods to determine the size of different glioblastoma regions of interest: the contrast-enhancing area, the necrotic area, the gross target volume and the volume of the edema area. The two methods, the ellipsoid formula (the calculated method) and the manual delineation (the measured method) showed a high correlation to determine glioblastoma volume and a high agreement to classify patients assessment response to treatment according to RANO criteria. This study revealed that calculated and measured methods could be used in clinical practice to estimate glioblastoma volume size and to evaluate tumor size evolution.
Image registration and segmentation are the two most studied problems in medical image analysis. Deep learning algorithms have recently gained a lot of attention due to their success and ...state-of-the-art results in variety of problems and communities. In this paper, we propose a novel, efficient, and multi-task algorithm that addresses the problems of image registration and brain tumor segmentation jointly. Our method exploits the dependencies between these tasks through a natural coupling of their interdependencies during inference. In particular, the similarity constraints are relaxed within the tumor regions using an efficient and relatively simple formulation. We evaluated the performance of our formulation both quantitatively and qualitatively for registration and segmentation problems on two publicly available datasets (BraTS 2018 and OASIS 3), reporting competitive results with other recent state-of-the-art methods. Moreover, our proposed framework reports significant amelioration (
< 0.005) for the registration performance inside the tumor locations, providing a generic method that does not need any predefined conditions (e.g., absence of abnormalities) about the volumes to be registered. Our implementation is publicly available online at https://github.com/TheoEst/joint_registration_tumor_segmentation.
Local recurrence in gynecological malignancies occurring in a previously irradiated field is a challenging clinical issue. The most frequent curative-intent treatment is salvage surgery. ...Reirradiation, using three-dimensional image-guided brachytherapy (3D-IGBT), might be a suitable alternative. We reviewed recent literature concerning 3D-IGBT for reirradiation in the context of local recurrences from gynecological malignancies.
We conducted a large-scale literature research, and 15 original studies, responding to our research criteria, were finally selected.
Local control rates ranged from 44% to 71.4% at 2-5 years, and overall survival rates ranged from 39.5% to 78% at 2-5 years. Grade ≥3 toxicities ranged from 1.7% to 50%, with only one study reporting a grade 5 event. Results in terms of outcome and toxicities were highly variable depending on studies. Several studies suggested that local control could be improved with 2 Gy equivalent doses >40 Gy.
IGBT appears to be a feasible alternative to salvage surgery in inoperable patients or patients refusing surgery, with an acceptable outcome for patients who have no other curative therapeutic options, however at a high cost of long-term grade ≥3 toxicities in some studies. We recommend that patients with local recurrence from gynecologic neoplasm occurring in previously irradiated fields should be referred to highly experienced expert centers. Centralization of data and large-scale multicentric international prospective trials are warranted. Efforts should be made to improve local control while limiting the risk of toxicities.
Immuno-radiotherapy may improve outcomes for patients with advanced solid tumors, although optimized combination modalities remain unclear. Here, we report the colorectal (CRC) cohort analysis from ...the SABR-PDL1 trial that evaluated the PD-L1 inhibitor atezolizumab in combination with stereotactic body radiation therapy (SBRT) in advanced cancer patients.
Eligible patients received atezolizumab 1200 mg every 3 weeks until progression or unmanageable toxicity, together with ablative SBRT delivered concurrently with the 2nd cycle (recommended dose of 45 Gy in 3 fractions, adapted upon normal tissue tolerance constraint). SBRT was delivered to at least one tumor site, with at least one additional measurable lesion being kept from the radiation field. The primary efficacy endpoint was one-year progression-free survival (PFS) rate from the start of atezolizumab. Sequential tumor biopsies were collected for deep multi-feature immune profiling.
Sixty pretreated (median of 2 prior lines) advanced CRC patients (38 men 63%; median age, 59 years range, 20-81 years; 77% with liver metastases) were enrolled in five centers (France: n = 4, Spain: n = 1) from 11/2016 to 04/2019. All but one (98%) received atezolizumab and 54/60 (90%) received SBRT. The most frequently irradiated site was lung (n = 30/54; 56.3%). Treatment-related G3 (no G4-5) toxicity was observed in 3 (5%) patients. Median OS and PFS were respectively 8.4 95%CI:5.9-11.6 and 1.4 months 95%CI:1.2-2.6, including five (9%) patients with PFS > 1 year (median time to progression: 19.2 months, including 2/5 MMR-proficient). Best overall responses consisted of stable disease (n = 38; 64%), partial (n = 3; 5%) and complete response (n = 1; 2%). Immune-centric multiplex IHC and RNAseq showed that SBRT redirected immune cells towards tumor lesions, even in the case of radio-induced lymphopenia. Baseline tumor PD-L1 and IRF1 nuclear expression (both in CD3 + T cells and in CD68 + cells) were higher in responding patients. Upregulation of genes that encode for proteins known to increase T and B cell trafficking to tumors (CCL19, CXCL9), migration (MACF1) and tumor cell killing (GZMB) correlated with responses.
This study provides new data on the feasibility, efficacy, and immune context of tumors that may help identifying advanced CRC patients most likely to respond to immuno-radiotherapy.
EudraCT N°: 2015-005464-42; Clinicaltrial.gov number: NCT02992912.
The integration of artificial intelligence (AI) and positron emission tomography (PET) imaging has the potential to become a powerful tool in drug discovery. This review aims to provide an overview ...of the current state of research and highlight the potential for this alliance to advance pharmaceutical innovation by accelerating the development and deployment of novel therapeutics. We previously performed a scoping review of three databases (Embase, MEDLINE, and CENTRAL), identifying 87 studies published between 2018 and 2022 relevant to medical imaging (e.g., CT, PET, MRI), immunotherapy, artificial intelligence, and radiomics. Herein, we reexamine the previously identified studies, performing a subgroup analysis on articles specifically utilizing AI and PET imaging for drug discovery purposes in immunotherapy-treated oncology patients. Of the 87 original studies identified, 15 met our updated search criteria. In these studies, radiomics features were primarily extracted from PET/CT images in combination (
= 9, 60.0%) rather than PET imaging alone (
= 6, 40.0%), and patient cohorts were mostly recruited retrospectively and from single institutions (
= 10, 66.7%). AI models were used primarily for prognostication (
= 6, 40.0%) or for assisting in tumor phenotyping (
= 4, 26.7%). About half of the studies stress-tested their models using validation sets (
= 4, 26.7%) or both validation sets and test sets (
= 4, 26.7%), while the remaining six studies (40.0%) either performed no validation at all or used less stringent methods such as cross-validation on the training set. Overall, the integration of AI and PET imaging represents a paradigm shift in drug discovery, offering new avenues for more efficient development of therapeutics. By leveraging AI algorithms and PET imaging analysis, researchers could gain deeper insights into disease mechanisms, identify new drug targets, or optimize treatment regimens. However, further research is needed to validate these findings and address challenges such as data standardization and algorithm robustness.
Strong rationale and a growing number of preclinical and clinical studies support combining radiotherapy and immunotherapy to improve patient outcomes. However, several critical questions remain, ...such as the identification of patients who will benefit from immunotherapy and the identification of the best modalities of treatment to optimize patient response. Imaging biomarkers and radiomics have recently emerged as promising tools for the non-invasive assessment of the whole disease of the patient, allowing comprehensive analysis of the tumor microenvironment, the spatial heterogeneity of the disease and its temporal changes. This review presents the potential applications of medical imaging and the challenges to address, in order to help clinicians choose the optimal modalities of both radiotherapy and immunotherapy, to predict patient’s outcomes and to assess response to these promising combinations.
Immunotherapy has greatly improved the outcomes of patients with metastatic melanoma. However, it has also led to new patterns of response and progression, creating an unmet need for better ...biomarkers to identify patients likely to achieve a lasting clinical benefit or experience immune-related adverse events. In this study, we performed a focused literature survey covering the application of artificial intelligence (AI; in the form of radiomics, machine learning, and deep learning) to patients diagnosed with melanoma and treated with immunotherapy, reviewing 12 studies relevant to the topic published up to early 2022. The most commonly investigated imaging modality was CT imaging in isolation (
= 9, 75.0%), while patient cohorts were most frequently recruited retrospectively and from single institutions (
= 7, 58.3%). Most studies concerned the development of AI tools to assist in prognostication (
= 5, 41.7%) or the prediction of treatment response (
= 6, 50.0%). Validation methods were disparate, with two studies (16.7%) performing no validation and equal numbers using cross-validation (
= 3, 25%), a validation set (
= 3, 25%), or a test set (
= 3, 25%). Only one study used both validation and test sets (
= 1, 8.3%). Overall, promising results have been observed for the application of AI to immunotherapy-treated melanoma. Further improvement and eventual integration into clinical practice may be achieved through the implementation of rigorous validation using heterogeneous, prospective patient cohorts.