Search

A significant challenge in machine learning-based medical image analysis is the scarcity of medical images. Obtaining a large number of labeled medical images is difficult because annotating medical images is a time-consuming process that requires specialized knowledge. In addition, inappropriate annotation processes can increase model bias. Self-supervised learning (SSL) is a type of unsupervised learning method that extracts image representations. Thus, SSL can be an effective method to reduce the number of labeled images. In this study, we investigated the feasibility of reducing the number of labeled images in a limited set of unlabeled medical images. The unlabeled chest X-ray (CXR) images were pretrained using the SimCLR framework, and then the representations were fine-tuned as supervised learning for the target task. A total of 2000 task-specific CXR images were used to perform binary classification of coronavirus disease 2019 (COVID-19) and normal cases. The results demonstrate that the performance of pretraining on task-specific unlabeled CXR images can be maintained when the number of labeled CXR images is reduced by approximately 40%. In addition, the performance was significantly better than that obtained without pretraining. In contrast, a large number of pretrained unlabeled images are required to maintain performance regardless of task specificity among a small number of labeled CXR images. In summary, to reduce the number of labeled images using SimCLR, we must consider both the number of images and the task-specific characteristics of the target images., (© 2024. The Author(s).)

Purpose: The purpose of this study was to assess the ability of pretreatment PET parameters and peripheral blood biomarkers to predict progression-free survival (PFS) and overall survival (OS) in NSCLC patients treated with ICIT., Methods: We prospectively included 87 patients in this study who underwent pre-treatment [ 18 F]-FDG PET/CT. Organ-specific and total metabolic tumor volume (MTV) and total lesion glycolysis (TLG) were measured using a semiautomatic software. Sites of organ involvement (SOI) were assessed by PET/CT. The log-rank test and Cox-regression analysis were used to assess associations between clinical, laboratory, and imaging parameters with PFS and OS. Time dependent ROC were calculated and model performance was evaluated in terms of its clinical utility., Results: MTV increased with the number of SOI and was correlated with neutrophil and lymphocyte cell count (Spearman's rho = 0.27 or 0.32; p =.02 or 0.003; respectively). Even after adjustment for known risk factors, such as PD-1 expression and neutrophil cell count, the MTV and the number of SOI were independent risk factors for progression (per 100 cm 3 ; adjusted hazard ratio [aHR]: 1.13; 95% confidence interval [95%CI]: 1.01-1.28; p =.04; single SOI vs. ≥ 4 SOI: aHR: 2.26, 95%CI: 1.04-4.94; p =.04). MTV and the number of SOI were independent risk factors for overall survival (per 100 cm 3 aHR: 1.11, 95%CI: 1.01-1.23; p =.03; single SOI vs. ≥ 4 SOI: aHR: 4.54, 95%CI: 1.64-12.58; p =.04). The combination of MTV and the number of SOI improved the risk stratification for PFS and OS (log-rank test p <.001; C-index: 0.64 and 0.67)., Conclusion: The MTV and the number of SOI are simple imaging markers that provide complementary information to facilitate risk stratification in NSCLC patients scheduled for ICIT., Competing Interests: Declaration of Competing Interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Helmut Prosch reports financial support was provided by MSD. Helmut Prosch reports financial support was provided by Roche. Helmut Prosch reports was provided by AstraZeneca Pharmaceuticals LP. Helmut Prosch reports was provided by Bristol Myers Squibb Co. Helmut Prosch reports financial support was provided by Christian Doppler Laboratories. Lucian Beer reports financial support was provided by Christian Doppler Laboratories., (Copyright © 2023. Published by Elsevier B.V.)

The introduction of neoadjuvant immune checkpoint inhibitors plus platinum-based chemotherapy has changed treatment regimens of patient's early-stage lung cancer. This treatment combination induces high rates of complete pathologic response and improves clinical endpoints. Imaging plays a fundamental role in assessment of treatment response, monitoring of (immune-related) adverse events and enables both the surgeon and pathologist optimal treatment and diagnostic workup of the resected tumor samples. Knowledge of the strengths and weaknesses of diagnostic imaging in this setting are essential for radiologists to provide valuable input in multidisciplinary team decisions., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper: [Helmut Prosch reports financial support was provided by Christian Doppler Laboratories. Lucian Beer reports a relationship with Bayer, Novartis, MSD, Roche that includes: speaking and lecture fees and travel reimbursement]., (Copyright © 2023 The Authors. Published by Elsevier B.V. All rights reserved.)

A wide spectrum of conditions, from life-threatening to non-urgent, can manifest with acute dyspnea, thus presenting major challenges for the treating physician when establishing the diagnosis and severity of the underlying disease. Imaging plays a decisive role in the assessment of acute dyspnea of cardiac and/or pulmonary origin. This article presents an overview of the current imaging modalities used to narrow the differential diagnosis in the assessment of acute dyspnea of cardiac or pulmonary origin. The current indications, findings, accuracy, and limits of each imaging modality are reported. Chest radiography is usually the primary imaging modality applied. There is a low radiation dose associated with this method, and it can assess the presence of fluid in the lung or pleura, consolidations, hyperinflation, pneumothorax, as well as heart enlargement. However, its low sensitivity limits the ability of the chest radiograph to accurately identify the causes of acute dyspnea. CT provides more detailed imaging of the cardiorespiratory system, and therefore, better sensitivity and specificity results, but it is accompanied by higher radiation exposure. Ultrasonography has the advantage of using no radiation, and is fast and feasible as a bedside test and appropriate for the assessment of unstable patients. However, patient-specific factors, such as body habitus, may limit its image quality and interpretability. Advances in knowledge This review provides guidance to the appropriate choice of imaging modalities in the diagnosis of patients with dyspnea of cardiac or pulmonary origin., Competing Interests: Competing interests: Ruxandra-Iulia Milos, Carmen Bartha, Benedikt H. Heidinger, Florian Prayer, Lucian Beer, Christian Wassipaul, Daria Kifjak, Martin Watzenböck – no conflict of interests to declare; Sebastian Röhrich – consulting activities contextflow GmbH; Helmut Prosch – Speaker honoria: AstraZeneca, BMS, Boehringer Ingelheim, Siemens Healthcare, Takeda. Research grants: Boehringer Ingelheim, (© 2023 The Authors. Published by the British Institute of Radiology.)