Machine learning and radiomics for segmentation and classification of adnexal masses on ultrasound

Ultrasound-based models exist to support the classification of adnexal masses but are subjective and rely upon ultrasound expertise. We aimed to develop an end-to-end machine learning (ML) model capable of automating the classification of adnexal masses. In this retrospective study, transvaginal ultrasound scan images with linked diagnoses (ultrasound subjective assessment or histology) were extracted and segmented from Imperial College Healthcare, UK (ICH development dataset; n = 577 masses; 1444 images) and Morgagni-Pierantoni Hospital, Italy (MPH external dataset; n = 184 masses; 476 images). A segmentation and classification model was developed using convolutional neural networks and traditional radiomics features. Dice surface coefficient (DICE) was used to measure segmentation performance and area under the ROC curve (AUC), F1-score and recall for classification performance. The ICH and MPH datasets had a median age of 45 (IQR 35-60) and 48 (IQR 38-57) years old and consisted of 23.1% and 31.5% malignant cases, respectively. The best segmentation model achieved a DICE score of 0.85 ± 0.01, 0.88 ± 0.01 and 0.85 ± 0.01 in the ICH training, ICH validation and MPH test sets. The best classification model achieved a recall of 1.00 and F1-score of 0.88 (AUC:0.93), 0.94 (AUC:0.89) and 0.83 (AUC:0.90) in the ICH training, ICH validation and MPH test sets, respectively. We have developed an end-to-end radiomics-based model capable of adnexal mass segmentation and classification, with a comparable predictive performance (AUC 0.90) to the published performance of expert subjective assessment (gold standard), and current risk models. Further prospective evaluation of the classification performance of this ML model against existing methods is required.

Levels of the cancer biomarker CA 19-9 are associated with thrombin generation in plasma from treatment-naïve pancreatic cancer patients

BACKGROUND

Pancreatic ductal adenocarcinoma (PDAC) is associated with a hypercoagulable state and high mortality. Increases in the plasma levels of tumor marker carbohydrate antigen (CA) 19-9 are used in diagnosis and follow-up but have also been reported to precede venous thromboembolism (VTE).

AIMS

We examined the association between CA 19-9 and thrombin generation (TG) in plasma from PDAC patients, as well as their association with coagulation biomarkers prior to pancreatic surgery. In addition, we determined the effect of commercial sources of CA 19-9 on TG.

METHODS

We collected plasma from 58 treatment-naïve PDAC patients without any signs of VTE. We measured levels of CA 19-9, FVIII, fibrinogen, D-dimer, antithrombin and extracellular vesicle (EV) tissue factor (TF) activity and TG using a Calibrated Automated Thrombogram (CAT). The effect of different commercial sources of CA 19-9 on TG in Standard Human Plasma (SHP) was also studied.

RESULTS

Patient plasma samples were divided into 4 preoperative groups based on the level of CA 19-9: none < 2, low = 3-200, high = 201-1000, and very high > 1000 U/mL. CA 19-9 levels were associated with several of the TG parameters, including endogenous thrombin potential, peak, and time to peak. CA 19-9 did not associate with any of the coagulation biomarkers. Spiking of SHP with CA 19-9 increased TG but this was decreased by an anti-TF antibody.

CONCLUSIONS

CA 19-9 was associated with TG in patients prior to any pancreatic cancer treatments or signs of VTE. Some commercial sources of CA 19-9 enhanced TG in SHP seemingly due to contaminating TF.

Dietary and urinary metabonomic factors possibly accounting for higher blood pressure of black compared with white Americans: results of International Collaborative Study on macro-/micronutrients and blood pressure

Black compared with non-Hispanic white Americans have higher systolic and diastolic blood pressure and rates of prehypertension/hypertension. Reasons for these adverse findings remain obscure. Analyses here focused on relations of foods/nutrients/urinary metabolites and higher black blood pressure for 369 black compared with 1190 non-Hispanic white Americans aged 40 to 59 years from 8 population samples. Multiple linear regression, standardized data from four 24-hour dietary recalls per person, two 24-hour urine collections, and 8 blood pressure measurements were used to quantitate the role of foods, nutrients, and metabolites in higher black blood pressure. Compared with non-Hispanic white Americans, blacks' average systolic/diastolic pressure was higher by 4.7/3.4 mm Hg (men) and 9.0/4.8 mm Hg (women). Control for higher body mass index of black women reduced excess black systolic/diastolic pressure to 6.8/3.8 mm Hg. Lesser intake of vegetables, fruits, grains, vegetable protein, glutamic acid, starch, fiber, minerals, and potassium, and higher intake of processed meats, pork, eggs, and sugar-sweetened beverages, along with higher cholesterol and higher Na/K ratio, related to in higher black blood pressure. Control for 11 nutrient and 10 non-nutrient correlates reduced higher black systolic/diastolic pressure to 2.3/2.3 mm Hg (52% and 33% reduction in men) and to 5.3/2.8 mm Hg (21% and 27% reduction in women). Control for foods/urinary metabolites had little further influence on higher black blood pressure. Less favorable multiple nutrient intake by blacks than non-Hispanic white Americans accounted, at least in part, for higher black blood pressure. Improved dietary patterns can contribute to prevention/control of more adverse black blood pressure levels.

Increased dispersion of refractoriness in the absence of QT prolongation in patients with mitral valve prolapse and ventricular arrhythmias

BACKGROUND

The mechanism responsible for the reported high incidence of ventricular arrhythmias in mitral valve prolapse is not clear. Electrocardiographic studies show an increased occurrence of repolarisation abnormalities on the 12 lead surface electrocardiogram, indicating regional differences in ventricular recovery. The purpose of this study was to investigate whether dispersion of refractoriness was an arrhythmogenic mechanism.

METHODS

QT dispersion was measured in 32 patients with echocardiographically documented mitral valve prolapse and ventricular arrhythmias on 24 hour Holter recordings. QT dispersion was defined as the difference between the maximum and minimum average QT interval in any of the 12 leads of the surface electrocardiogram. QT dispersion corrected for heart rate was calculated by Bazett's formula. The results were compared with the data from 32 matched controls without a history of cardiac disease. Patients taking drugs that influence the QT interval and patients with a QRS duration > 120 ms were excluded.

RESULTS

QT dispersion was greater in patients with mitral valve prolapse than in matched controls (60 (20) v 39 (11 ms) respectively, P < or = 0.001) as was corrected QT (64 (20 ms) v 43 (12 ms) respectively, P < or = 0.001). There was no significant difference in minimum or maximum QT intervals between the two groups.

CONCLUSIONS

QT dispersion on the 12 lead surface electrocardiogram was greater in patients with mitral valve prolapse with ventricular arrhythmias than in normal controls, but the maximum QT interval was not increased. The results accord with the hypothesis that regional shortening and lengthening of repolarisation times in patients with mitral valve prolapse may account for the increased dispersion of refractoriness.