To estimate cumulative radiation exposure and lifetime attributable risk (LAR) of radiation-induced cancer from computed tomographic (CT) scanning of adult patients at a tertiary care academic medical center. This HIPAA-compliant study was approved by the institutional review board with waiver of informed consent. The cohort comprised 31,462 patients who underwent diagnostic CT in 2007 and had undergone 190,712 CT examinations over the prior 22 years. Each patient's cumulative CT radiation exposure was estimated by summing typical CT effective doses, and the Biological Effects of Ionizing Radiation (BEIR) VII methodology was used to estimate LAR on the basis of sex and age at each exposure. Billing ICD9 codes and electronic order entry information were used to stratify patients with LAR greater than 1%. Thirty-three percent of patients underwent five or more lifetime CT examinations, and 5% underwent between 22 and 132 examinations. Fifteen percent received estimated cumulative effective doses of more than 100 mSv, and 4% received between 250 and 1375 mSv. Associated LAR had mean and maximum values of 0.3% and 12% for cancer incidence and 0.2% and 6.8% for cancer mortality, respectively. CT exposures were estimated to produce 0.7% of total expected baseline cancer incidence and 1% of total cancer mortality. Seven percent of the cohort had estimated LAR greater than 1%, of which 40% had either no malignancy history or a cancer history without evidence of residual disease. Cumulative CT radiation exposure added incrementally to baseline cancer risk in the cohort. While most patients accrue low radiation-induced cancer risks, a subgroup is potentially at higher risk due to recurrent CT imaging.

The RSNA Pediatric Bone Age Machine Learning Challenge showed the application of machine learning in medical imaging, promoted ways in which these emerging tools and methods can improve diagnostic care, and identified innovators in machine learning applications in medical imaging. PurposeThe Radiological Society of North America (RSNA) Pediatric Bone Age Machine Learning Challenge was created to show an application of machine learning (ML) and artificial intelligence (AI) in medical imaging, promote collaboration to catalyze AI model creation, and identify innovators in medical imaging.Materials and MethodsThe goal of this challenge was to solicit individuals and teams to create an algorithm or model using ML techniques that would accurately determine skeletal age in a curated data set of pediatric hand radiographs. The primary evaluation measure was the mean absolute distance (MAD) in months, which was calculated as the mean of the absolute values of the difference between the model estimates and those of the reference standard, bone age.ResultsA data set consisting of 14 236 hand radiographs (12 611 training set, 1425 validation set, 200 test set) was made available to registered challenge participants. A total of 260 individuals or teams registered on the Challenge website. A to...

This dataset is intended to be used for machine learning and is composed of annotations with bounding boxes for pulmonary opacity on chest radiographs which may represent pneumonia in the appropriate clinical setting.

OBJECTIVE. The purposes of this article are to describe concepts that radiologists should understand to evaluate machine learning projects, including common algorithms, supervised as opposed to unsupervised techniques, statistical pitfalls, and data considerations for training and evaluation, and to briefly describe ethical dilemmas and legal risk. CONCLUSION. Machine learning includes a broad class of computer programs that improve with experience. The complexity of creating, training, and monitoring machine learning indicates that the success of the algorithms will require radiologist involvement for years to come, leading to engagement rather than replacement.

Artificial intelligence using deep learning algorithms shows promise in detecting emergent neurologic disease at noncontrast-enhanced head CT, which supports further evaluation of the algorithms within a controlled and prospective clinical setting to determine whether it can independently screen noncontrast-enhanced head CT examinations and alert the interpreting radiologist of a potential critical finding. PurposeTo evaluate the performance of an artificial intelligence (AI) tool using a deep learning algorithm for detecting hemorrhage, mass effect, or hydrocephalus (HMH) at non—contrast material–enhanced head computed tomographic (CT) examinations and to determine algorithm performance for detection of suspected acute infarct (SAI).Materials and MethodsThis HIPAA-compliant retrospective study was completed after institutional review board approval. A training and validation dataset of noncontrast-enhanced head CT examinations that comprised 100 examinations of HMH, 22 of SAI, and 124 of noncritical findings was obtained resulting in 2583 representative images. Examinations were processed by using a convolutional neural network (deep learning) using two different window and level configurations (brain window and stroke window). AI algorithm performance was tested on a separate dataset containing 50 examinations with HMH findings, 15 with SAI findings, and 35 with noncritical findings.ResultsFinal algorithm ...