Machine learning analysis of breast ultrasound to classify triple negative and HER2+ breast cancer subtypes.

OBJECTIVES: Early diagnosis of triple-negative (TN) and human epidermal growth factor receptor 2 positive (HER2+) breast cancer is important due to its increased risk of micrometastatic spread necessitating early treatment and for guiding targeted therapies. This study aimed to evaluate the diagnostic performance of machine learning (ML) classification of newly diagnosed breast masses into TN versus non-TN (NTN) and HER2+ versus HER2 negative (HER2-) breast cancer, using radiomic features extracted from grayscale ultrasound (US) b-mode images. MATERIALS AND METHODS: A retrospective chart review identified 88 female patients who underwent diagnostic breast US imaging, had confirmation of invasive malignancy on pathology and receptor status determined on immunohistochemistry available. The patients were classified as TN, NTN, HER2+ or HER2- for ground-truth labelling. For image analysis, breast masses were manually segmented by a breast radiologist. Radiomic features were extracted per image and used for predictive modelling. Supervised ML classifiers included: logistic regression, k-nearest neighbour, and Naïve Bayes. Classification performance measures were calculated on an independent (unseen) test set. The area under the receiver operating characteristic curve (AUC), sensitivity (%), and specificity (%) were reported for each classifier. RESULTS: The logistic regression classifier demonstrated the highest AUC: 0.824 (sensitivity: 81.8%, specificity: 74.2%) for the TN sub-group and 0.778 (sensitivity: 71.4%, specificity: 71.6%) for the HER2 sub-group. CONCLUSION: ML classifiers demonstrate high diagnostic accuracy in classifying TN versus NTN and HER2+ versus HER2- breast cancers using US images. Identification of more aggressive breast cancer subtypes early in the diagnostic process could help achieve better prognoses by prioritizing clinical referral and prompting adequate early treatment.

Signs of post-traumatic hypovolemia on abdominal CT and their clinical importance: A systematic review.

PURPOSE: Our aim was to assess the findings of hypovolemia on abdominal CT that are most frequently seen in blunt abdominal trauma patients. When possible, we assessed the correlation of these CT signs with clinical outcome. METHODS: MEDLINE, CENTRAL and EMBASE were systematically searched. Two reviewers independently screened and included articles and performed the data-extraction. Primary outcomes of interest were the frequency of each sign and its correlation with mortality. Secondary outcomes were need for intervention, transfusion need, intensive care unit admission rate and length of stay. RESULTS: A flat inferior vena cava and an inferior vena cava halo, a diminished aortic calibre, shock bowel, altered enhancement of the liver, pancreas, adrenals, kidneys, spleen and gallbladder, peripancreatic fluid and splenic volume changes have been described in the setting of hypovolemic trauma patients to constellate a CT hypovolemic shock complex. It is argued that vascular signs represent the true hypovolemic state and the visceral signs represent hypoperfusion. There is no consensus on the frequency or clinical relevance of these signs, which at least partly can be explained by the heterogeneity in study design, study population, scanning protocols and outcome parameters. Available evidence suggests a good predictive value for occult shock and a higher mortality rate when a flat inferior vena cava is present. Evidence regarding the other signs is scarce. CONCLUSIONS: The hypovolemic shock complex is an entity of both vascular and visceral CT signs that can be seen in blunt trauma patients. It can offer guidance to a swift primary imaging survey in the acute trauma setting, allowing the radiologist to alert the treating physicians to possible pending hypovolemic shock.