The natural language processing of radiology requests and reports of chest imaging: Comparing five transformer models' multilabel classification and a proof-of-concept study.

BACKGROUND: Radiology requests and reports contain valuable information about diagnostic findings and indications, and transformer-based language models are promising for more accurate text classification. METHODS: In a retrospective study, 2256 radiologist-annotated radiology requests (8 classes) and reports (10 classes) were divided into training and testing datasets (90% and 10%, respectively) and used to train 32 models. Performance metrics were compared by model type (LSTM, Bertje, RobBERT, BERT-clinical, BERT-multilingual, BERT-base), text length, data prevalence, and training strategy. The best models were used to predict the remaining 40,873 cases' categories of the datasets of requests and reports. RESULTS: The RobBERT model performed the best after 4000 training iterations, resulting in AUC values ranging from 0.808 [95% CI (0.757-0.859)] to 0.976 [95% CI (0.956-0.996)] for the requests and 0.746 [95% CI (0.689-0.802)] to 1.0 [95% CI (1.0-1.0)] for the reports. The AUC for the classification of normal reports was 0.95 [95% CI (0.922-0.979)]. The predicted data demonstrated variability of both diagnostic yield for various request classes and request patterns related to COVID-19 hospital admission data. CONCLUSION: Transformer-based natural language processing is feasible for the multilabel classification of chest imaging request and report items. Diagnostic yield varies with the information in the requests.

Using deep learning to safely exclude lesions with only ultrafast breast MRI to shorten acquisition and reading time.

OBJECTIVES: To investigate the feasibility of automatically identifying normal scans in ultrafast breast MRI with artificial intelligence (AI) to increase efficiency and reduce workload. METHODS: In this retrospective analysis, 837 breast MRI examinations performed on 438 women from April 2016 to October 2019 were included. The left and right breasts in each examination were labelled normal (without suspicious lesions) or abnormal (with suspicious lesions) based on final interpretation. Maximum intensity projection (MIP) images of each breast were then used to train a deep learning model. A high sensitivity threshold was calculated based on the detection trade - off (DET) curve on the validation set. The performance of the model was evaluated by receiver operating characteristic analysis of the independent test set. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) with the high sensitivity threshold were calculated. RESULTS: The independent test set consisted of 178 examinations of 149 patients (mean age, 44 years ± 14 [standard deviation]). The trained model achieved an AUC of 0.81 (95% CI: 0.75-0.88) on the independent test set. Applying a threshold of 0.25 yielded a sensitivity of 98% (95% CI: 90%; 100%), an NPV of 98% (95% CI: 89%; 100%), a workload reduction of 15.7%, and a scan time reduction of 16.6%. CONCLUSION: This deep learning model has a high potential to help identify normal scans in ultrafast breast MRI and thereby reduce radiologists' workload and scan time. KEY POINTS: • Deep learning in TWIST may eliminate the necessity of additional sequences for identifying normal breasts during MRI screening. • Workload and scanning time reductions of 15.7% and 16.6%, respectively, could be achieved with the cost of 1 (1 of 55) false negative prediction.

Effects of control temperature, ablation time, and background tissue in radiofrequency ablation of osteoid osteoma: A computer modeling study.

To study the effects of the control temperature, ablation time, and the background tissue surrounding the tumor on the size of the ablation zone on radiofrequency ablation (RFA) of osteoid osteoma (OO). Finite element models of non-cooled temperature-controlled RFA of typical OOs were developed to determine the resulting ablation radius at control temperatures of 70, 80, and 90°C. Three different geometries were used, mimicking common cases of OO. The ablation radius was obtained by using the Arrhenius equation to determine cell viability. Ablation radii were larger for higher temperatures and also increased with time. All geometries and control temperatures tested had ablation radii larger than the tumor. The ablation radius developed rapidly in the first few minutes for all geometries and control temperatures tested, developing slowly towards the end of the ablation. Resistive heating and the temperature distribution showed differences depending on background tissue properties, resulting in differences in the ablation radius on each geometry. The ablation radius has a clear dependency not only on the properties of the tumor but also on the background tissue. Lower background tissue's electrical conductivity and blood perfusion rates seem to result in larger ablation zones. The differences observed between the different geometries suggest the need for patient-specific planning, as the anatomical variations could cause significantly different outcomes where models like the one here presented could help to guarantee safe and successful tumor ablations.

Deep convolutional neural networks for multiplanar lung nodule detection: Improvement in small nodule identification.

PURPOSE: Early detection of lung cancer is of importance since it can increase patients' chances of survival. To detect nodules accurately during screening, radiologists would commonly take the axial, coronal, and sagittal planes into account, rather than solely the axial plane in clinical evaluation. Inspired by clinical work, the paper aims to develop an accurate deep learning framework for nodule detection by a combination of multiple planes. METHODS: The nodule detection system is designed in two stages, multiplanar nodule candidate detection, multiscale false positive (FP) reduction. At the first stage, a deeply supervised encoder-decoder network is trained by axial, coronal, and sagittal slices for the candidate detection task. All possible nodule candidates from the three different planes are merged. To further refine results, a three-dimensional multiscale dense convolutional neural network that extracts multiscale contextual information is applied to remove non-nodules. In the public LIDC-IDRI dataset, 888 computed tomography scans with 1186 nodules accepted by at least three of four radiologists are selected to train and evaluate our proposed system via a tenfold cross-validation scheme. The free-response receiver operating characteristic curve is used for performance assessment. RESULTS: The proposed system achieves a sensitivity of 94.2% with 1.0 FP/scan and a sensitivity of 96.0% with 2.0 FPs/scan. Although it is difficult to detect small nodules (i.e., <6 mm), our designed CAD system reaches a sensitivity of 93.4% (95.0%) of these small nodules at an overall FP rate of 1.0 (2.0) FPs/scan. At the nodule candidate detection stage, results show that the system with a multiplanar method is capable to detect more nodules compared to using a single plane. CONCLUSION: Our approach achieves good performance not only for small nodules but also for large lesions on this dataset. This demonstrates the effectiveness of our developed CAD system for lung nodule detection.

Spin transport in insulators without exchange stiffness.

The discovery of new materials that efficiently transmit spin currents has been important for spintronics and material science. The electric insulator Gd3Ga5O12 (GGG), a standard substrate for growing magnetic films, can be a spin current generator, but has never been considered as a superior conduit for spin currents. Here we report spin current propagation in paramagnetic GGG over several microns. Surprisingly, spin transport persists up to temperatures of 100 K [Formula: see text] Tg = 180 mK, the magnetic glass-like transition temperature of GGG. At 5 K and 3.5 T, we find a spin diffusion length λGGG = 1.8 ± 0.2 µm and a spin conductivity σGGG = (7.3 ± 0.3) × 104 Sm-1 that is larger than that of the record quality magnet Y3Fe5O12 (YIG). We conclude that exchange stiffness is not required for efficient spin transport, which challenges conventional models and provides new material-design strategies for spintronic devices.

Formation and electronic properties of InSb nanocrosses.

Signatures of Majorana fermions have recently been reported from measurements on hybrid superconductor-semiconductor nanowire devices. Majorana fermions are predicted to obey special quantum statistics, known as non-Abelian statistics. To probe this requires an exchange operation, in which two Majorana fermions are moved around one another, which requires at least a simple network of nanowires. Here, we report on the synthesis and electrical characterization of crosses of InSb nanowires. The InSb wires grow horizontally on flexible vertical stems, allowing nearby wires to meet and merge. In this way, near-planar single-crystalline nanocrosses are created, which can be measured by four electrical contacts. Our transport measurements show that the favourable properties of the InSb nanowire devices-high carrier mobility and the ability to induce superconductivity--are preserved in the cross devices. Our nanocrosses thus represent a promising system for the exchange of Majorana fermions.