Outcome Prediction in Patients with Severe Traumatic Brain Injury Using Deep Learning from Head CT Scans.

Background After severe traumatic brain injury (sTBI), physicians use long-term prognostication to guide acute clinical care yet struggle to predict outcomes in comatose patients. Purpose To develop and evaluate a prognostic model combining deep learning of head CT scans and clinical information to predict long-term outcomes after sTBI. Materials and Methods This was a retrospective analysis of two prospectively collected databases. The model-building set included 537 patients (mean age, 40 years ± 17 [SD]; 422 men) from one institution from November 2002 to December 2018. Transfer learning and curriculum learning were applied to a convolutional neural network using admission head CT to predict mortality and unfavorable outcomes (Glasgow Outcomes Scale scores 1-3) at 6 months. This was combined with clinical input for a holistic fusion model. The models were evaluated using an independent internal test set and an external cohort of 220 patients with sTBI (mean age, 39 years ± 17; 166 men) from 18 institutions in the Transforming Research and Clinical Knowledge in Traumatic Brain Injury (TRACK-TBI) study from February 2014 to April 2018. The models were compared with the International Mission on Prognosis and Analysis of Clinical Trials in TBI (IMPACT) model and the predictions of three neurosurgeons. Area under the receiver operating characteristic curve (AUC) was used as the main model performance metric. Results The fusion model had higher AUCs than did the IMPACT model in the prediction of mortality (AUC, 0.92 [95% CI: 0.86, 0.97] vs 0.80 [95% CI: 0.71, 0.88]; P < .001) and unfavorable outcomes (AUC, 0.88 [95% CI: 0.82, 0.94] vs 0.82 [95% CI: 0.75, 0.90]; P = .04) on the internal data set. For external TRACK-TBI testing, there was no evidence of a significant difference in the performance of any models compared with the IMPACT model (AUC, 0.83; 95% CI: 0.77, 0.90) in the prediction of mortality. The Imaging model (AUC, 0.73; 95% CI: 0.66-0.81; P = .02) and the fusion model (AUC, 0.68; 95% CI: 0.60, 0.76; P = .02) underperformed as compared with the IMPACT model (AUC, 0.83; 95% CI: 0.77, 0.89) in the prediction of unfavorable outcomes. The fusion model outperformed the predictions of the neurosurgeons. Conclusion A deep learning model of head CT and clinical information can be used to predict 6-month outcomes after severe traumatic brain injury. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Haller in this issue.

Localization of Four Allergens in Artemisia Pollen by Immunofluorescent Antibodies.

BACKGROUND: Artemisia pollens have a high potential to induce allergic symptoms. Seven allergen components have been identified, but only Art v 7 has been localized in the pollen grain. This study aimed to localize the allergens in the pollen grains of 4 Artemisia spp. METHODS: Pollen extracts from 2 Chinese Artemisia spp., A. argyi and A. annua, were used to immunize BALB/c mice. Recombinant Art v 1 and Art v 3 allergens were used to select specific monoclonal antibodies (mAbs). Three mAbs were used to purify the natural allergens and were then analyzed by mass spectrometry. As reported previously, polyclonal antibodies were obtained from rabbits immunized with 3 synthesized peptides of Art an 7. Using conventional histology procedures with pollens from 4 Artemisia spp. (A. argyi, A. annua, A. capilaris, and A. sieversiana), allergen images were observed and recorded by fluorescence and confocal laser microscopy. RESULTS: We obtained 2 specific mAbs against Art v 1, 1 against Art v 2, and 4 against Art v 3 homologs. The Art v 1 and Art v 3 homologs were mainly located on the pollen walls, and the Art v 7 homologous protein was localized intracellularly around nuclei. The location of the Art v 2 homologous protein varied across species, being intracellular around nuclei for A. annua and A. argyi, and in both the pollen wall and around nuclei for A. capilaris and A. sieversiana. CONCLUSIONS: Four mugwort allergens were localized in the pollen, and the major Art v 1 and Art v 3 allergens were located mainly in the pollen wall.

Artemisia pollen allergy in China: Component-resolved diagnosis reveals allergic asthma patients have significant multiple allergen sensitization.

BACKGROUND: Artemisia pollen allergy is a major cause of asthma in Northern China. Possible associations between IgE responses to Artemisia allergen components and clinical phenotypes have not yet been evaluated. This study was to establish sensitization patterns of four Artemisia allergens and possible associations with demographic characteristics and clinical phenotypes in three areas of China. METHODS: Two hundred and forty patients allergic to Artemisia pollen were examined, 178 from Shanxi and 30 from Shandong Provinces in Northern China, and 32 from Yunnan Province in Southwestern China. Allergic asthma, rhinitis, conjunctivitis, and eczema symptoms were diagnosed. All patients' sera were tested by ImmunoCAP with mugwort pollen extract and the natural components nArt v 1, nArt ar 2, nArt v 3, and nArt an 7. RESULTS: The frequency of sensitization and the IgE levels of the four components in Artemisia allergic patients from Southwestern China were significantly lower than in those from the North. Art v 1 and Art an 7 were the most frequently recognized allergens (84% and 87%, respectively), followed by Art v 3 (66%) and Art ar 2 (48%). Patients from Northern China were more likely to have allergic asthma (50%) than patients from Southwestern China (3%), and being sensitized to more than two allergens increased the risk of allergic asthma, in which co-sensitization to three major allergens Art v 1, Art v 3, and Art an 7 is prominent. CONCLUSIONS: Component-resolved diagnosis of Chinese Artemisia pollen-allergic patients helps assess the potential risk of mugwort-associated allergic asthma.