Transformer-based language models adapted for radiology had higher performance on three radiology natural language processing tasks than five intensively studied baseline models. Purpose To investigate if tailoring a transformer-based language model to radiology is beneficial for radiology natural language processing (NLP) applications. Materials and Methods This retrospect...

The ability to accurately and non-invasively quantify highly concentrated magnetic nanoparticles (MNPs) is desirable for many emerging applications. Ultrashort echo time quantitative susceptibility mapping (UTE-QSM) has demonstrated the capability to detect high iron concentrations. In this study, we aimed to investigate the effect of different sampling trajectories on the accuracy of quantification based on MNPs acquired through UTE-QSM. A phantom with six different MNP concentrations was prepared for UTE-QSM study with different UTE sampling trajectories, including radial acquisition, continuous single point imaging (CSPI), and Cones with four different gradient stretching factors of 1.0, 1.2, 1.4, and 1.6. No significant differences were found in QSM values derived from the different UTE sampling strategies, suggesting that the UTE-QSM technique could be accelerated with extended Cones sampling.

Foot and ankle fractures are the most common military health problem. Automated diagnosis can save time and personnel. It is crucial to distinguish fractures not only from normal healthy cases, but also robust against the presence of other orthopedic pathologies. Artificial intelligence (AI) deep learning has been shown to be promising. Previously, we have developed HAMIL-Net to automatically detect orthopedic injuries for upper extremity injuries. In this research, we investigated the performance of HAMIL-Net for detecting foot and ankle fractures in the presence of other abnormalities. HAMIL-Net is a novel deep neural network consisting of a hierarchical attention layer followed by a multiple-instance learning layer. The design allowed it to deal with imaging studies with multiple views. We used 148K musculoskeletal imaging studies for 51K Veterans at VA San Diego in the past 20 years to create datasets for this research. We annotated each study by a semi-automated pipeline leveraging radiology reports written by board-certified radiologists and extracting findings with a natural language processing tool and manually validated the annotations. HAMIL-Net can be trained with study-level, multiple-view examples, and detect foot and ankle fractures with a 0.87 area under the receiver operational curve, but the performance dropped when tested by cases including other abnormalities. By integrating a fracture specialized model with one that detecting a broad range of abnormalities, HAMIL-Net’s accuracy of detecting any abnormality improved from 0.53 to 0.77 and F-score from 0.46 to 0.86. We also reported HAMIL-Net’s performance under different study types including for young (age 18–35) patients. Automated fracture detection is promising but to be deployed in clinical use, presence of other abnormalities must be considered to deliver its full benefit. Our results with HAMIL-Net showed that considering other abnormalities improved fracture detection and allowed for incidental findings of other musculoskeletal abnormalities pertinent or superimposed on fractures.

To study how the fat affects the quantification of magnetic susceptibility of iron oxide nanoparticles (IONPs) according to the chemical shift effects, especially at ultra-short echo times, simulation experiments were carried out in this paper. Numerical phantom of IONPs solutions with six different concentrations was designed. Magnetic resonance imaging signals with different fat fractions (ffs) were generated according to a multi-peak chemical shift model. Quantitative susceptibility mapping (QSM) was calculated, and the relationship between the QSM value and ff, as well as echo selection, is evaluated. The results show that streaking artifacts become more and more severe when ff increases, which makes QSM value overestimated. Shortened echo spacing time can reduce streaking artifacts with the same ff. These findings will benefit future clinical evaluations for quantification of iron in vivo.