Lung Nodule Classification Using Biomarkers, Volumetric Radiomics, and 3D CNNs.

We present a hybrid algorithm to estimate lung nodule malignancy that combines imaging biomarkers from Radiologist's annotation with image classification of CT scans. Our algorithm employs a 3D Convolutional Neural Network (CNN) as well as a Random Forest in order to combine CT imagery with biomarker annotation and volumetric radiomic features. We analyze and compare the performance of the algorithm using only imagery, only biomarkers, combined imagery + biomarkers, combined imagery + volumetric radiomic features, and finally the combination of imagery + biomarkers + volumetric features in order to classify the suspicion level of nodule malignancy. The National Cancer Institute (NCI) Lung Image Database Consortium (LIDC) IDRI dataset is used to train and evaluate the classification task. We show that the incorporation of semi-supervised learning by means of K-Nearest-Neighbors (KNN) can increase the available training sample size of the LIDC-IDRI, thereby further improving the accuracy of malignancy estimation of most of the models tested although there is no significant improvement with the use of KNN semi-supervised learning if image classification with CNNs and volumetric features is combined with descriptive biomarkers. Unexpectedly, we also show that a model using image biomarkers alone is more accurate than one that combines biomarkers with volumetric radiomics, 3D CNNs, and semi-supervised learning. We discuss the possibility that this result may be influenced by cognitive bias in LIDC-IDRI because malignancy estimates were recorded by the same radiologist panel as biomarkers, as well as future work to incorporate pathology information over a subset of study participants.

A methodology to estimate the number of unsafe vehicle-cyclist passing events on urban arterials.

In this study interactions between motorized vehicles and bicycles were studied by analyzing the overtaking behavior of motorized vehicles when passing bicycles on urban arterials. A methodology is presented to estimate the number of 'unsafe' passing events on 4-lane urban arterials with no on-street bike lanes. A 'critical passing distance' is defined to classify expected passing maneuvers i.e. when a motorized vehicle overtakes a bicycle, into 'safe' and 'unsafe' passing events. The proposed method enables calculation of the expected number of 'unsafe passing' events based on the expected bicycle demand, road segment's length, AADT, speed limit, and traffic signal timing parameters. The 'critical passing distance' is an input parameter and can be set by the planner. Given the number of expected 'unsafe passing' events, and institutional safety objectives and standards in terms of acceptable risk levels for cyclists, transportation planning departments can use the proposed methodology to decide whether provision of a specific cycling facility is necessary for a given road segment.

Current Challenges in the Use of Computed Tomography and MR Imaging in Suspected Cervical Spine Trauma.

There is controversy regarding the optimal imaging strategy in adult blunt trauma patients for suspected cervical spine trauma. Some investigators recommend negative computed tomography (CT) alone to clear the cervical spine in adult blunt trauma patients, while others insist that MR imaging is necessary, especially among obtunded adult blunt trauma patients. CT is an excellent imaging modality for bony cervical spine injury; however, there is a nonzero rate of clinically significant cervical spine injuries missed on CT. MR imaging has high sensitivity for soft tissue cervical spine injuries, but low specificity for the rare isolated unstable ligamentous cervical spine injury.