HaN-Seg: The head and neck organ-at-risk CT and MR segmentation challenge.

BACKGROUND AND PURPOSE: To promote the development of auto-segmentation methods for head and neck (HaN) radiation treatment (RT) planning that exploit the information of computed tomography (CT) and magnetic resonance (MR) imaging modalities, we organized HaN-Seg: The Head and Neck Organ-at-Risk CT and MR Segmentation Challenge. MATERIALS AND METHODS: The challenge task was to automatically segment 30 organs-at-risk (OARs) of the HaN region in 14 withheld test cases given the availability of 42 publicly available training cases. Each case consisted of one contrast-enhanced CT and one T1-weighted MR image of the HaN region of the same patient, with up to 30 corresponding reference OAR delineation masks. The performance was evaluated in terms of the Dice similarity coefficient (DSC) and 95-percentile Hausdorff distance (HD95), and statistical ranking was applied for each metric by pairwise comparison of the submitted methods using the Wilcoxon signed-rank test. RESULTS: While 23 teams registered for the challenge, only seven submitted their methods for the final phase. The top-performing team achieved a DSC of 76.9 % and a HD95 of 3.5 mm. All participating teams utilized architectures based on U-Net, with the winning team leveraging rigid MR to CT registration combined with network entry-level concatenation of both modalities. CONCLUSION: This challenge simulated a real-world clinical scenario by providing non-registered MR and CT images with varying fields-of-view and voxel sizes. Remarkably, the top-performing teams achieved segmentation performance surpassing the inter-observer agreement on the same dataset. These results set a benchmark for future research on this publicly available dataset and on paired multi-modal image segmentation in general.

High prevalence of pre-existing sarcopenia in critically ill patients with hematologic malignancies admitted to the intensive care unit for sepsis or septic shock.

BACKGROUND & AIMS: We aimed to evaluate body composition (BC) by computed tomography (CT) in hematologic malignancy (HM) patients admitted to the intensive care unit (ICU) for sepsis or septic shock. METHODS: We retrospectively assessed BC and its impact on outcome of 186 patients at the 3rd lumbar (L3) and 12th thoracic vertebral levels (T12) using CT-scan performed before ICU admission. RESULTS: The median patient age was 58.0 [47; 69] years. Patients displayed adverse clinical characteristics at admission with median [q1; q3] SAPS II and SOFA scores of 52 [40; 66] and 8 [5; 12], respectively. The mortality rate in the ICU was 45.7%. Overall survival rates at 1 month after admission in the pre-existing sarcopenic vs. non pre-existing sarcopenic patients were 47.9% (95% CI [37.6; 61.0]) and 55.0% (95% CI [41.6; 72.8]), p = 0.99), respectively, at the L3 level and 48.4% (95% CI [40.4; 58.0]) vs. 66.7% (95% CI [51.1; 87.0]), p = 0.062), respectively, at the T12 level. CONCLUSIONS: Sarcopenia is assessable by CT scan at both the T12 and L3 levels and is highly prevalent in HM patients admitted to the ICU for severe infections. Sarcopenia may contribute to the high mortality rate in the ICU in this population.

Prognostic relevance of sarcopenia, geriatric, and nutritional assessments in older patients with diffuse large B-cell lymphoma: results of a multicentric prospective cohort study.

This prospective study aimed to investigate the prognostic effect of sarcopenia, geriatric, and nutritional status in older patients with diffuse large B-cell lymphoma (DLBCL). Ninety-five patients with DLBCL older than 70 years who were treated with immunochemotherapy were included. The lumbar L3 skeletal muscle index (L3-SMI) was measured by computed tomography at baseline, and sarcopenia was defined as low L3-SMI. Geriatric assessment included G8 score, CIRS-G scale, Timed Up and Go test, and instrumental activity of daily living. Nutritional status was assessed using the Mini Nutritional Assessment and the body mass index, and several scores used in the literature incorporating nutritional and inflammatory biomarkers, namely the Nutritional and inflammatory status (NIS), Geriatric Nutritional Risk Index, Prognostic Nutritional Index, and Glasgow Prognostic Score.Fifty-three patients were considered sarcopenic. Sarcopenic patients displayed higher levels of inflammation markers and lower levels of prealbumin than non-sarcopenic patients. Sarcopenia was associated with NIS, but was not associated with severe adverse events and treatment disruptions. They were, however, more frequent among patients with elevated NIS. Sarcopenia did not appear in this study as a prognostic factor for progression-free survival (PFS) or overall survival (OS). However, NIS emerged as predictive of the outcome with a 2-year PFS rate of 88% in the NIS ≤ 1 group and 49% in the NIS > 1 group and a significant effect in a multivariate analysis for both PFS (p = 0.049) and OS (HR = 9.61, CI 95% = [1.03-89.66], p = 0.04). Sarcopenia was not associated with adverse outcomes, but was related to NIS, which appeared to be an independent prognostic factor.

Prediction of brain tumor recurrence location based on multi-modal fusion and nonlinear correlation learning.

Brain tumor is one of the leading causes of cancer death. The high-grade brain tumors are easier to recurrent even after standard treatment. Therefore, developing a method to predict brain tumor recurrence location plays an important role in the treatment planning and it can potentially prolong patient's survival time. There is still little work to deal with this issue. In this paper, we present a deep learning-based brain tumor recurrence location prediction network. Since the dataset is usually small, we propose to use transfer learning to improve the prediction. We first train a multi-modal brain tumor segmentation network on the public dataset BraTS 2021. Then, the pre-trained encoder is transferred to our private dataset for extracting the rich semantic features. Following that, a multi-scale multi-channel feature fusion model and a nonlinear correlation learning module are developed to learn the effective features. The correlation between multi-channel features is modeled by a nonlinear equation. To measure the similarity between the distributions of original features of one modality and the estimated correlated features of another modality, we propose to use Kullback-Leibler divergence. Based on this divergence, a correlation loss function is designed to maximize the similarity between the two feature distributions. Finally, two decoders are constructed to jointly segment the present brain tumor and predict its future tumor recurrence location. To the best of our knowledge, this is the first work that can segment the present tumor and at the same time predict future tumor recurrence location, making the treatment planning more efficient and precise. The experimental results demonstrated the effectiveness of our proposed method to predict the brain tumor recurrence location from the limited dataset.

Quantitative MRI to Characterize Hypoxic Tumors in Comparison to FMISO PET/CT for Radiotherapy in Oropharynx Cancers.

Intratumoral hypoxia is associated with a poor prognosis and poor response to treatment in head and neck cancers. Its identification would allow for increasing the radiation dose to hypoxic tumor subvolumes. 18F-FMISO PET imaging is the gold standard; however, quantitative multiparametric MRI could show the presence of intratumoral hypoxia. Thus, 16 patients were prospectively included and underwent 18F-FDG PET/CT, 18F-FMISO PET/CT, and multiparametric quantitative MRI (DCE, diffusion and relaxometry T1 and T2 techniques) in the same position before treatment. PET and MRI sub-volumes were segmented and classified as hypoxic or non-hypoxic volumes to compare quantitative MRI parameters between normoxic and hypoxic volumes. In total, 13 patients had hypoxic lesions. The Dice, Jaccard, and overlap fraction similarity indices were 0.43, 0.28, and 0.71, respectively, between the FDG PET and MRI-measured lesion volumes, showing that the FDG PET tumor volume is partially contained within the MRI tumor volume. The results showed significant differences in the parameters of SUV in FDG and FMISO PET between patients with and without measurable hypoxic lesions. The quantitative MRI parameters of ADC, T1 max mapping and T2 max mapping were different between hypoxic and normoxic subvolumes. Quantitative MRI, based on free water diffusion and T1 and T2 mapping, seems to be able to identify intra-tumoral hypoxic sub-volumes for additional radiotherapy doses.

Assessing Interobserver Variability in the Delineation of Structures in Radiation Oncology: A Systematic Review.

PURPOSE: The delineation of target volumes and organs at risk is the main source of uncertainty in radiation therapy. Numerous interobserver variability (IOV) studies have been conducted, often with unclear methodology and nonstandardized reporting. We aimed to identify the parameters chosen in conducting delineation IOV studies and assess their performances and limits. METHODS AND MATERIALS: We conducted a systematic literature review to highlight major points of heterogeneity and missing data in IOV studies published between 2018 and 2021. For the main used metrics, we did in silico analyses to assess their limits in specific clinical situations. RESULTS: All disease sites were represented in the 66 studies examined. Organs at risk were studied independently of tumor site in 29% of reviewed IOV studies. In 65% of studies, statistical analyses were performed. No gold standard (GS; ie, reference) was defined in 36% of studies. A single expert was considered as the GS in 21% of studies, without testing intraobserver variability. All studies reported both absolute and relative indices, including the Dice similarity coefficient (DSC) in 68% and the Hausdorff distance (HD) in 42%. Limitations were shown in silico for small structures when using the DSC and dependence on irregular shapes when using the HD. Variations in DSC values were large between studies, and their thresholds were inconsistent. Most studies (51%) included 1 to 10 cases. The median number of observers or experts was 7 (range, 2-35). The intraclass correlation coefficient was reported in only 9% of cases. Investigating the feasibility of studying IOV in delineation, a minimum of 8 observers with 3 cases, or 11 observers with 2 cases, was required to demonstrate moderate reproducibility. CONCLUSIONS: Implementation of future IOV studies would benefit from a more standardized methodology: clear definitions of the gold standard and metrics and a justification of the tradeoffs made in the choice of the number of observers and number of delineated cases should be provided.

Evaluation of sarcopenia biomarkers in older patients undergoing major surgery for digestive cancer. SAXO prospective cohort study.

BACKGROUND: The aim of the study was to prospectively evaluate different biomarkers to identify the most reliable for anticipating complications after major abdominal surgery for digestive cancer in older patients and compare their performance to the existing definition and screening algorithm of sarcopenia from EWGSOP. METHODS: Ninety-five consecutive patients aged over 65 years who underwent elective surgery for digestive cancer were prospectively included in the SAXO study. Sarcopenia was defined according to EWGSOP criteria (four level from no sarcopenia to severe sarcopenia). Strength and physical performance were evaluated with the handgrip test (HGT) and gait speed test (GST), respectively. CT scan analysis was used to calculate the skeletal muscle index (SMI), intermuscular adipose tissue (IMAT), visceral adipose tissue (VAT), and subcutaneous adipose tissue (SAT). Measures were adjusted to body mass index (BMI). Complication grading was performed using the Clavien‒Dindo classification. A doubly robust estimator with multivariable regression was used to limit bias. RESULTS: Sixteen patients presented with sarcopenia. Adjusted to BMI, sarcopenic patients had an increased IMATBMI (0.35 vs. 0.22; p = 0.003) and increased VATBMI (7.85 vs. 6.13; p = 0.048). In multivariable analysis, IMAT was an independent risk factor for minor and severe complications (OR = 1.298; 95% CI [1.031: 1.635] p = 0.027), while an increased SAT area was a protective factor (OR = 0.982; 95% CI [0.969: 0.995] p = 0.007). Twenty-two patients were obese (BMI ≥30 kg/m2). While no association was observed between obesity and sarcopenia (according to EWGSOP definition), obese patients had increased IMATBMI (0.31 vs. 0.23; p = 0.010) and VATBMI (8.40 vs. 6.49; p = 0.019). The combination of SAT, VAT and IMAT performed well to anticipate severe complication (AUC = 0.759) while AUC of EWGSOP 2010 and 2019 algorithm were 0.660 and 0.519, respectively. DISCUSSION: Non-invasive and imaging related measures of IMAT, SAT and VAT seems to be valuable tools to refine risk-assessment of older patients in surgery and specially to detect myosteatosis in obese ones.

Multi-task multi-scale learning for outcome prediction in 3D PET images.

BACKGROUND AND OBJECTIVES: Predicting patient response to treatment and survival in oncology is a prominent way towards precision medicine. To this end, radiomics has been proposed as a field of study where images are used instead of invasive methods. The first step in radiomic analysis in oncology is lesion segmentation. However, this task is time consuming and can be physician subjective. Automated tools based on supervised deep learning have made great progress in helping physicians. However, they are data hungry, and annotated data remains a major issue in the medical field where only a small subset of annotated images are available. METHODS: In this work, we propose a multi-task, multi-scale learning framework to predict patient's survival and response. We show that the encoder can leverage multiple tasks to extract meaningful and powerful features that improve radiomic performance. We also show that subsidiary tasks serve as an inductive bias so that the model can better generalize. RESULTS: Our model was tested and validated for treatment response and survival in esophageal and lung cancers, with an area under the ROC curve of 77% and 71% respectively, outperforming single-task learning methods. CONCLUSIONS: Multi-task multi-scale learning enables higher performance of radiomic analysis by extracting rich information from intratumoral and peritumoral regions.

Correction: Brochet et al. A Quantitative Comparison between Shannon and Tsallis-Havrda-Charvat Entropies Applied to Cancer Outcome Prediction. Entropy 2022, 24, 436.

Alexandre Huat, Sébastien Thureau, David Pasquier, Isabelle Gardin, Romain Modzelewski, David Gibon, Juliette Thariat and Vincent Grégoire were not included as authors in the original publication [...].

Clinical and phantom validation of a deep learning based denoising algorithm for F-18-FDG PET images from lower detection counting in comparison with the standard acquisition.

BACKGROUND: PET/CT image quality is directly influenced by the F-18-FDG injected activity. The higher the injected activity, the less noise in the reconstructed images but the more radioactive staff exposition. A new FDA cleared software has been introduced to obtain clinical PET images, acquired at 25% of the count statistics considering US practices. Our aim is to determine the limits of a deep learning based denoising algorithm (SubtlePET) applied to statistically reduced PET raw data from 3 different last generation PET scanners in comparison to the regular acquisition in phantom and patients, considering the European guidelines for radiotracer injection activities. Images of low and high contrasted (SBR = 2 and 5) spheres of the IEC phantom and high contrast (SBR = 5) of micro-spheres of Jaszczak phantom were acquired on 3 different PET devices. 110 patients with different pathologies were included. The data was acquired in list-mode and retrospectively reconstructed with the regular acquisition count statistic (PET100), 50% reduction in counts (PET50) and 66% reduction in counts (PET33). These count reduced images were post-processed with SubtlePET to obtain PET50 + SP and PET33 + SP images. Patient image quality was scored by 2 senior nuclear physicians. Peak-signal-to-Noise and Structural similarity metrics were computed to compare the low count images to regular acquisition (PET100). RESULTS: SubtlePET reliably denoised the images and maintained the SUVmax values in PET50 + SP. SubtlePET enhanced images (PET33 + SP) had slightly increased noise compared to PET100 and could lead to a potential loss of information in terms of lesion detectability. Regarding the patient datasets, the PET100 and PET50 + SP were qualitatively comparable. The SubtlePET algorithm was able to correctly recover the SUVmax values of the lesions and maintain a noise level equivalent to full-time images. CONCLUSION: Based on our results, SubtlePET is adapted in clinical practice for half-time or half-dose acquisitions based on European recommended injected dose of 3 MBq/kg without diagnostic confidence loss.

Weakly Supervised Tumor Detection in PET Using Class Response for Treatment Outcome Prediction.

It is proven that radiomic characteristics extracted from the tumor region are predictive. The first step in radiomic analysis is the segmentation of the lesion. However, this task is time consuming and requires a highly trained physician. This process could be automated using computer-aided detection (CAD) tools. Current state-of-the-art methods are trained in a supervised learning setting, which requires a lot of data that are usually not available in the medical imaging field. The challenge is to train one model to segment different types of tumors with only a weak segmentation ground truth. In this work, we propose a prediction framework including a 3D tumor segmentation in positron emission tomography (PET) images, based on a weakly supervised deep learning method, and an outcome prediction based on a 3D-CNN classifier applied to the segmented tumor regions. The key step is to locate the tumor in 3D. We propose to (1) calculate two maximum intensity projection (MIP) images from 3D PET images in two directions, (2) classify the MIP images into different types of cancers, (3) generate the class activation maps through a multitask learning approach with a weak prior knowledge, and (4) segment the 3D tumor region from the two 2D activation maps with a proposed new loss function for the multitask. The proposed approach achieves state-of-the-art prediction results with a small data set and with a weak segmentation ground truth. Our model was tested and validated for treatment response and survival in lung and esophageal cancers on 195 patients, with an area under the receiver operating characteristic curve (AUC) of 67% and 59%, respectively, and a dice coefficient of 73% and 0.77% for tumor segmentation.

A Quantitative Comparison between Shannon and Tsallis-Havrda-Charvat Entropies Applied to Cancer Outcome Prediction.

In this paper, we propose to quantitatively compare loss functions based on parameterized Tsallis-Havrda-Charvat entropy and classical Shannon entropy for the training of a deep network in the case of small datasets which are usually encountered in medical applications. Shannon cross-entropy is widely used as a loss function for most neural networks applied to the segmentation, classification and detection of images. Shannon entropy is a particular case of Tsallis-Havrda-Charvat entropy. In this work, we compare these two entropies through a medical application for predicting recurrence in patients with head-neck and lung cancers after treatment. Based on both CT images and patient information, a multitask deep neural network is proposed to perform a recurrence prediction task using cross-entropy as a loss function and an image reconstruction task. Tsallis-Havrda-Charvat cross-entropy is a parameterized cross-entropy with the parameter α. Shannon entropy is a particular case of Tsallis-Havrda-Charvat entropy for α=1. The influence of this parameter on the final prediction results is studied. In this paper, the experiments are conducted on two datasets including in total 580 patients, of whom 434 suffered from head-neck cancers and 146 from lung cancers. The results show that Tsallis-Havrda-Charvat entropy can achieve better performance in terms of prediction accuracy with some values of α.

Deep learning analysis of contrast-enhanced spectral mammography to determine histoprognostic factors of malignant breast tumours.

OBJECTIVE: To evaluate if a deep learning model can be used to characterise breast cancers on contrast-enhanced spectral mammography (CESM). METHODS: This retrospective mono-centric study included biopsy-proven invasive cancers with an enhancement on CESM. CESM images include low-energy images (LE) comparable to digital mammography and dual-energy subtracted images (DES) showing tumour angiogenesis. For each lesion, histologic type, tumour grade, estrogen receptor (ER) status, progesterone receptor (PR) status, HER-2 status, Ki-67 proliferation index, and the size of the invasive tumour were retrieved. The deep learning model used was a CheXNet-based model fine-tuned on CESM dataset. The area under the curve (AUC) of the receiver operating characteristic (ROC) curve was calculated for the different models: images by images and then by majority voting combining all the incidences for one tumour. RESULTS: In total, 447 invasive breast cancers detected on CESM with pathological evidence, in 389 patients, which represented 2460 images analysed, were included. Concerning the ER, the deep learning model on the DES images had an AUC of 0.83 with the image-by-image analysis and of 0.85 for the majority voting. For the triple-negative analysis, a high AUC was observable for all models, in particularity for the model on LE images with an AUC of 0.90 for the image-by-image analysis and 0.91 for the majority voting. The AUC for the other histoprognostic factors was lower. CONCLUSION: Deep learning analysis on CESM has the potential to determine histoprognostic tumours makers, notably estrogen receptor status, and triple-negative receptor status. KEY POINTS: • A deep learning model developed for chest radiography was adapted by fine-tuning to be used on contrast-enhanced spectral mammography. • The adapted models allowed to determine for invasive breast cancers the status of estrogen receptors and triple-negative receptors. • Such models applied to contrast-enhanced spectral mammography could provide rapid prognostic and predictive information.

First Comparison between [18f]-FMISO and [18f]-Faza for Preoperative Pet Imaging of Hypoxia in Lung Cancer.

Hypoxic areas are typically resistant to treatment. However, the fluorine-18-fluoroazomycin-arabinoside (FAZA) and fluorine 18 misonidazole (FMISO) tracers have never been compared in non small cell lung cancer (NSCLC). This study compares the capability of 18F-FAZA PET/CT with that of 18F-FMISO PET/CT for detecting hypoxic tumour regions in early and locally advanced NSCLC patients. We prospectively evaluated patients who underwent preoperative PET scans before surgery for localised NSCLC (i.e., fluorodeoxyglucose (FDG)-PET, FMISO-PET, and FAZA-PET). The PET data of the three tracers were compared with each other and then compared to immunohistochemical analysis (GLUT-1, CAIX, LDH-5, and HIF1-Alpha) after tumour resection. Overall, 19 patients with a mean age of 68.2 ± 8 years were included. There were 18 lesions with significant uptake (i.e., SUVmax >1.4) for the F-MISO and 17 for FAZA. The mean SUVmax was 3 (±1.4) with a mean volume of 25.8 cc (±25.8) for FMISO and 2.2 (±0.7) with a mean volume of 13.06 cc (±13.76) for FAZA. The SUVmax of F-MISO was greater than that of FAZA (p = 0.0003). The SUVmax of F-MISO shows a good correlation with that of FAZA at 0.86 (0.66-0.94). Immunohistochemical results are not correlated to hypoxia PET regardless of the staining. The two tracers show a good correlation with hypoxia, with FMISO being superior to FAZA. FMISO, therefore, remains the reference tracer for defining hypoxic volumes.

Targeting the Urotensin II/UT G Protein-Coupled Receptor to Counteract Angiogenesis and Mesenchymal Hypoxia/Necrosis in Glioblastoma.

Glioblastomas (GBMs) are the most common primary brain tumors characterized by strong invasiveness and angiogenesis. GBM cells and microenvironment secrete angiogenic factors and also express chemoattractant G protein-coupled receptors (GPCRs) to their advantage. We investigated the role of the vasoactive peptide urotensin II (UII) and its receptor UT on GBM angiogenesis and tested potential ligand/therapeutic options based on this system. On glioma patient samples, the expression of UII and UT increased with the grade with marked expression in the vascular and peri-necrotic mesenchymal hypoxic areas being correlated with vascular density. In vitro human UII stimulated human endothelial HUV-EC-C and hCMEC/D3 cell motility and tubulogenesis. In mouse-transplanted Matrigel sponges, mouse (mUII) and human UII markedly stimulated invasion by macrophages, endothelial, and smooth muscle cells. In U87 GBM xenografts expressing UII and UT in the glial and vascular compartments, UII accelerated tumor development, favored hypoxia and necrosis associated with increased proliferation (Ki67), and induced metalloproteinase (MMP)-2 and -9 expression in Nude mice. UII also promoted a "tortuous" vascular collagen-IV expressing network and integrin expression mainly in the vascular compartment. GBM angiogenesis and integrin αvß3 were confirmed by in vivo 99mTc-RGD tracer imaging and tumoral capture in the non-necrotic area of U87 xenografts in Nude mice. Peptide analogs of UII and UT antagonist were also tested as potential tumor repressor. Urotensin II-related peptide URP inhibited angiogenesis in vitro and failed to attract vascular and inflammatory components in Matrigel in vivo. Interestingly, the UT antagonist/biased ligand urantide and the non-peptide UT antagonist palosuran prevented UII-induced tubulogenesis in vitro and significantly delayed tumor growth in vivo. Urantide drastically prevented endogenous and UII-induced GBM angiogenesis, MMP, and integrin activations, associated with GBM tumoral growth. These findings show that UII induces GBM aggressiveness with necrosis and angiogenesis through integrin activation, a mesenchymal behavior that can be targeted by UT biased ligands/antagonists.

Evaluation of an Automatic Classification Algorithm Using Convolutional Neural Networks in Oncological Positron Emission Tomography.

Introduction: Our aim was to evaluate the performance in clinical research and in clinical routine of a research prototype, called positron emission tomography (PET) Assisted Reporting System (PARS) (Siemens Healthineers) and based on a convolutional neural network (CNN), which is designed to detect suspected cancer sites in fluorine-18 fluorodeoxyglucose (18F-FDG) PET/computed tomography (CT). Method: We retrospectively studied two cohorts of patients. The first cohort consisted of research-based patients who underwent PET scans as part of the initial workup for diffuse large B-cell lymphoma (DLBCL). The second cohort consisted of patients who underwent PET scans as part of the evaluation of miscellaneous cancers in clinical routine. In both cohorts, we assessed the correlation between manually and automatically segmented total metabolic tumor volumes (TMTVs), and the overlap between both segmentations (Dice score). For the research cohort, we also compared the prognostic value for progression-free survival (PFS) and overall survival (OS) of manually and automatically obtained TMTVs. Results: For the first cohort (research cohort), data from 119 patients were retrospectively analyzed. The median Dice score between automatic and manual segmentations was 0.65. The intraclass correlation coefficient between automatically and manually obtained TMTVs was 0.68. Both TMTV results were predictive of PFS (hazard ratio: 2.1 and 3.3 for automatically based and manually based TMTVs, respectively) and OS (hazard ratio: 2.4 and 3.1 for automatically based and manually based TMTVs, respectively). For the second cohort (routine cohort), data from 430 patients were retrospectively analyzed. The median Dice score between automatic and manual segmentations was 0.48. The intraclass correlation coefficient between automatically and manually obtained TMTVs was 0.61. Conclusion: The TMTVs determined for the research cohort remain predictive of total and PFS for DLBCL. However, the segmentations and TMTVs determined automatically by the algorithm need to be verified and, sometimes, corrected to be similar to the manual segmentation.

Multi-task deep learning based CT imaging analysis for COVID-19 pneumonia: Classification and segmentation.

This paper presents an automatic classification segmentation tool for helping screening COVID-19 pneumonia using chest CT imaging. The segmented lesions can help to assess the severity of pneumonia and follow-up the patients. In this work, we propose a new multitask deep learning model to jointly identify COVID-19 patient and segment COVID-19 lesion from chest CT images. Three learning tasks: segmentation, classification and reconstruction are jointly performed with different datasets. Our motivation is on the one hand to leverage useful information contained in multiple related tasks to improve both segmentation and classification performances, and on the other hand to deal with the problems of small data because each task can have a relatively small dataset. Our architecture is composed of a common encoder for disentangled feature representation with three tasks, and two decoders and a multi-layer perceptron for reconstruction, segmentation and classification respectively. The proposed model is evaluated and compared with other image segmentation techniques using a dataset of 1369 patients including 449 patients with COVID-19, 425 normal ones, 98 with lung cancer and 397 of different kinds of pathology. The obtained results show very encouraging performance of our method with a dice coefficient higher than 0.88 for the segmentation and an area under the ROC curve higher than 97% for the classification.

Prognostic value of sarcopenia in patients treated by Radiochemotherapy for locally advanced oesophageal cancer.

BACKGROUND: Sarcopenia is defined by a loss of skeletal muscle mass with or without loss of fat mass. Sarcopenia has been associated to reduced tolerance to treatment and worse prognosis in cancer patients, including patients undergoing surgery for limited oesophageal cancer. Concomitant chemo-radiotherapy is the standard treatment for locally-advanced tumour, not accessible to surgical resection. Using automated delineation of the skeletal muscle, we have investigated the prognostic value of sarcopenia in locally advanced oesophageal cancer (LAOC) patients treated by curative-intent chemo-radiotherapy. METHODS: The clinical, nutritional, anthropometric, and functional-imaging (18FDG-PET/CT) data were collected in 97 patients treated between 2006 and 2012 in our institution. The skeletal muscle area was automatically delineated on cross-sectional CT images acquired at the 3rd. lumbar vertebra level and divided by the patient's squared height (SML3/h2) to obtain the Skeletal Muscle Index (SMI). The primary endpoint was overall survival probability. RESULTS: Seventy-six deaths were reported. The median survival time was 27 [95% Confidence Interval 23-40] months for the whole population. Univariate analyses (Cox Proportional Hazard Model) showed decreased survival probabilities in patients with reduced SMI, WHO > 0, Body Mass Index ≤21, and Nutritional Risk Index ≤97.5. Multivariate analyses showed that sarcopenia was the only significant prognostic factor (HR 2.32 [1.24-4.34], p = 0.008). Using Receiver Operating Characteristics curves, the Area Under the Curve (AUC) was 0.73 in males (p = 0.0002], the optimal threshold being 51.5 cm2/m2. In women, the AUC was 0.65 (p = 0.19). CONCLUSION: Sarcopenia is a powerful independent prognostic factor, associated with a rise of the overall mortality in patients treated exclusively by radiochemotherapy for a locally advanced oesophageal cancer. L3 CT images are easily gathered from 18FDG-PET/CT acquisitions.

Development of Novel 111-In-Labelled DOTA Urotensin II Analogues for Targeting the UT Receptor Overexpressed in Solid Tumours.

Overexpression of G protein-coupled receptors (GPCRs) in tumours is widely used to develop GPCR-targeting radioligands for solid tumour imaging in the context of diagnosis and even treatment. The human vasoactive neuropeptide urotensin II (hUII), which shares structural analogies with somatostatin, interacts with a single high affinity GPCR named UT. High expression of UT has been reported in several types of human solid tumours from lung, gut, prostate, or breast, suggesting that UT is a valuable novel target to design radiolabelled hUII analogues for cancer diagnosis. In this study, two original urotensinergic analogues were first conjugated to a DOTA chelator via an aminohexanoic acid (Ahx) hydrocarbon linker and then -hUII and DOTA-urantide, complexed to the radioactive metal indium isotope to successfully lead to radiolabelled DOTA-Ahx-hUII and DOTA-Ahx-urantide. The 111In-DOTA-hUII in human plasma revealed that only 30% of the radioligand was degraded after a 3-h period. DOTA-hUII and DOTA-urantide exhibited similar binding affinities as native peptides and relayed calcium mobilization in HEK293 cells expressing recombinant human UT. DOTA-hUII, not DOTA-urantide, was able to promote UT internalization in UT-expressing HEK293 cells, thus indicating that radiolabelled 111In-DOTA-hUII would allow sufficient retention of radioactivity within tumour cells or radiolabelled DOTA-urantide may lead to a persistent binding on UT at the plasma membrane. The potential of these radioligands as candidates to target UT was investigated in adenocarcinoma. We showed that hUII stimulated the migration and proliferation of both human lung A549 and colorectal DLD-1 adenocarcinoma cell lines endogenously expressing UT. In vivo intravenous injection of 111In-DOTA-hUII in C57BL/6 mice revealed modest organ signals, with important retention in kidney. 111In-DOTA-hUII or 111In-DOTA-urantide were also injected in nude mice bearing heterotopic xenografts of lung A549 cells or colorectal DLD-1 cells both expressing UT. The observed significant renal uptake and low tumour/muscle ratio (around 2.5) suggest fast tracer clearance from the organism. Together, DOTA-hUII and DOTA-urantide were successfully radiolabelled with 111Indium, the first one functioning as a UT agonist and the second one as a UT-biased ligand/antagonist. To allow tumour-specific targeting and prolong body distribution in preclinical models bearing some solid tumours, these radiolabelled urotensinergic analogues should be optimized for being used as potential molecular tools for diagnosis imaging or even treatment tools.