Comparing Radiomics features of tumour and healthy liver tissue in a limited CT dataset: A machine learning study

IntroductionLiver cancer lesions on Computed Tomography (CT) withholds a great amount of data, which is not visible to the radiologists and radiographer. Radiomics features can be extracted from the lesions and used to train Machine Learning (ML) algorithms to predict between tumour and liver tissue. The purpose of this study was to investigate and classify Radiomics features extracted from liver tumours and normal liver tissue in a limited CT dataset.MethodsThe Liver Tumour Segmentation Benchmark (LiTS) dataset consisting of 131 CT scans of the liver with segmentations of tumour tissue and healthy liver was used to extract Radiomic features. Extracted Radiomic features included size, shape, and location extracted with morphological and statistical techniques according to the International Symposium on Biomedical Imaging manual. Relevant features was selected with chi2 correlation and principal component analysis (PCA) with tumour and healthy liver tissue as outcome according to a consensus between three experienced radiologists. Logistic regression, random forest and support vector machine was used to train and validate the dataset with a 10-fold cross-validation method and the Grid Search as hyper-parameter tuning. Performance was evaluated with sensitivity, specificity and accuracy.ResultsThe performance of the ML algorithms achieved sensitivities, specificities and accuracy ranging from 96.30% (95% CI: 81.03%–99.91%) to 100.00% (95% CI: 86.77%–100.00%), 91.30% (95% CI: 71.96%–98.93%) to 100.00% (95% CI: 83.89%–100.00%)and 94.00% (95% CI: 83.45%–98.75%) to 100.00% (95% CI: 92.45%–100.00%), respectively.ConclusionML algorithms classifies Radiomics features extracted from healthy liver and tumour tissue with perfect accuracy. The Radiomics signature allows for a prognostic biomarker for hepatic tumour screening on liver CT.Implications for practiceDifferentiation between tumour and liver tissue with Radiomics ML algorithms have the potential to increase the diagnostic accuracy, assist in the decision-making of supplementary multiphasic enhanced medical imaging, as well as for developing novel prognostic biomarkers for liver cancer patients.     

肝细胞癌TACE术后病灶存活或复发18F-FDGPET/CT和增强CT诊断价值Meta分析

目的采用Meta分析方法评价¹⁸F-脱氧葡萄糖正电子发射计算机体层摄影(¹⁸F-FDG PET/CT)和增强CT(CECT)诊断经导管肝动脉化疗栓塞术(TACE)术后存活或复发病灶的临床价值。方法根据PRISMA报告规范开展Meta分析。检索PubMed、Embase、Cochrane Library、Web of Science、中国知网、万方和维普数据库中¹⁸F-FDG PET/CT和CECT诊断TACE术后存活或复发病灶的临床研究,时间至2019-04。由2位研究人员独立筛选文献、提取资料,根据诊断准确性研究质量评价工具-2(QUADAS-2)评价纳入研究的偏倚风险后,采用Stata 12.0软件进行Meta分析,计算其汇总敏感度(Sen)和特异度(Spe),绘制受试者工作特征曲线(SROC)并计算曲线下面积(AUC)。结果共纳入10篇¹⁸F-FDG PET/CT及13篇CECT诊断TACE术后存活或复发病灶的原始研究,分别包括322例患者的467个病灶和748例患者的943个病灶。Meta分析显示,¹⁸F-FDG PET/CT诊断TACE术后存活或复发病灶的Sen=0.92(95%CI为0.87~0.94)、Spe=0.95(95%CI为0.82~0.99)、AUC=0.97(95%CI为0.93~0.99);CECT诊断TACE术后存活或复发病灶的Sen=0.72(95%CI为0.66~0.78)、Spe=0.99(95%CI为0.93~1.00)、AUC=0.87(95%CI为0.83~0.89)。此外,CECT诊断TACE术后存活或复发Sen(Z=2.34,P=0.02)和AUC(Z=2.21,P=0.03)值低于¹⁸F-FDG PET/CT,差异有统计学意义。结论相比于CECT,¹⁸F-FDG PET/CT对TACE术后存活或复发病灶具有较高诊断效能,可视为TACE术后存活或复发病灶有效的影像学诊断方法。     

Patient Preferences for Benefits,Risks, and Administration Route of Hypomethylating Agents in Myelodysplastic Syndromes

Introduction/BackgroundTherapy with infused or injected hypomethylating agents (HMAs) may lead to higher treatment administration burden (ie, local reaction, visit frequency and duration) vs. oral HMAs. ObjectivesTo reveal preferences of US and Canadian patients with myelodysplastic syndromes (MDS) for HMAs’ benefits, risks, and administration burden through an online discrete-choice experiment (DCE).Materials and MethodsChoice of DCE attributes and survey development were informed by literature review and interviews with clinicians, MDS patients, and caregivers serving as patient proxies, and patient advocacy groups (PAGs) representatives, including from AAMAC, AAMDS, and MDSF. DCE choice tasks were analyzed using random parameter logit models. Survey patients were recruited by the PAGs via their networks. To understand key preference drivers and how much patients were willing to trade between attributes, we calculated each attribute's relative attribute importance (RAI) and marginal rates of substitution.ResultsOne hundred eighty-four respondents (including 158 patients; mean age, 67.2 years; male, 50.5%; White, 50.5%; US residents, 88%) completed the survey. MDS risk was low (34.8%), high (30.9%), or unknown (34.2%). RAI (in decreasing order) was as follows: risk of AML (40%), fatigue level (33%), number of visits (12%), mode of administration (6%), visit duration (5%), and administration frequency (4%). Assuming the same risk of AML transformation or level of fatigue, most respondents (76.6%) were predicted to switch to an oral pill if it were available to them.ConclusionGiven equivalent effectiveness across HMAs, patients’ preferences for HMA administration method should be considered in treatment decision-making to minimize burden and facilitate adherence.     

Anticoagulation Strategies in Patients With Cancer: JACC Review Topic of the Week

Patients with active cancer are at an increased risk of arterial and venous thromboembolism (VTE) and bleeding events. Historically, in patients with cancer, low molecular weight heparins have been preferred for treatment of VTE, whereas warfarin has been the standard anticoagulant for stroke prevention in patients with atrial fibrillation (AF). More recently, direct oral anticoagulants (DOACs) have been demonstrated to reduce the risk of venous and arterial thromboembolism in large randomized clinical trials of patients with VTE and AF, respectively, thus providing an attractive oral dosing option that does not require routine laboratory monitoring. In this review, we summarize available clinical trial data and guideline recommendations, and outline a practical approach to anticoagulation management of VTE and AF in cancer.     

Venetoclax in AML: Where We Are and Where We Are Headed

The developmental path for venetoclax in acute myeloid leukemia (AML) has been rapid and stands in stark contrast to the incremental progress that has characterized the field in previous decades. For perspective, on December 31, 2013, the first AML patient was enrolled into a study using venetoclax; 59 months later, on November 21, 2018, venetoclax received accelerated approval by the FDA for use in AML. In June 2020, Dr. DiNardo presented the results of the required confirmatory study at the European Hematology Association meeting, showing that venetoclax with azacitidine resulted in a superior response rate and overall survival compared to azacitidine alone for older, newly diagnosed AML patients. This swift progress has provided a welcome and potent new therapy for patients with AML; with it come questions about how its role can be expanded, and how its use can be optimized.