Brain magnetic resonance imaging (MRI) is useful for predicting the outcome of patients with acute ischemic stroke (AIS). Although deep learning (DL) using brain MRI with certain image biomarkers has shown satisfactory results in predicting poor outcomes, no study has assessed the usefulness of natural language processing (NLP)-based machine learning (ML) algorithms using brain MRI free-text reports of AIS patients. Therefore, we aimed to assess whether NLP-based ML algorithms using brain MRI text reports could predict poor outcomes in AIS patients. This study included only English text reports of brain MRIs examined during admission of AIS patients. Poor outcome was defined as a modified Rankin Scale score of 3–6, and the data were captured by trained nurses and physicians. We only included MRI text report of the first MRI scan during the admission. The text dataset was randomly divided into a training and test dataset with a 7:3 ratio. Text was vectorized to word, sentence, and document levels. In the word level approach, which did not consider the sequence of words, and the “bag-of-words” model was used to reflect the number of repetitions of text token. The “sent2vec” method was used in the sensation-level approach considering the sequence of words, and the word embedding was used in the document level approach. In addition to conventional ML algorithms, DL algorithms such as the convolutional neural network (CNN), long short-term memory, and multilayer perceptron were used to predict poor outcomes using 5-fold cross-validation and grid search techniques. The performance of each ML classifier was compared with the area under the receiver operating characteristic (AUROC) curve. Among 1840 subjects with AIS, 645 patients (35.1%) had a poor outcome 3 months after the stroke onset. Random forest was the best classifier (0.782 of AUROC) using a word-level approach. Overall, the document-level approach exhibited better performance than did the word- or sentence-level approaches. Among all the ML classifiers, the multi-CNN algorithm demonstrated the best classification performance (0.805), followed by the CNN (0.799) algorithm. When predicting future clinical outcomes using NLP-based ML of radiology free-text reports of brain MRI, DL algorithms showed superior performance over the other ML algorithms. In particular, the prediction of poor outcomes in document-level NLP DL was improved more by multi-CNN and CNN than by recurrent neural network-based algorithms. NLP-based DL algorithms can be used as an important digital marker for unstructured electronic health record data DL prediction.

Park H, et al. Clin Exp Otorhinolaryngol. 2015 Jun;8(2):92-96. http://dx.doi.org/10.3342/ceo.2015.8.2.92

Conclusion: Seven-day administration of systemic steroids was more effective in preserving hearing for 12 weeks after cochlear implantation (CI) than a 3-day delivery. Objectives: To determine the effectiveness of extended delivery of systemic steroids to preserve hearing in guinea pigs after CI. Methods: Dexamethasone (4 mg/ml) was delivered parenterally via a mini-osmotic pump for either 3 or 7 days. A dummy CI electrode was inserted via cochleostomy approach in 8-week-old guinea pigs. Auditory thresholds were assessed from tone burst auditory brainstem responses (2, 8, 16, 24, and 32 kHz) at 1 day prior to CI, and 1, 4, and 12 weeks after implantation. Histologic evaluation of the cochleae was carried out. Results: No differences were observed in hearing thresholds among groups before CI. Significant hearing preservation was achieved at 8, 16, 24, and 32 kHz only in the 7-day infusion group compared with the control group at 1 week after CI. The same trend was maintained at 4 weeks (16, 24 kHz) and 12 weeks (16, 24, and 32 kHz). Histologic review of the 7-day infusion group revealed less fibrosis and ossification in the scala tympani and the preservation of more spiral ganglion cells, compared with the control group.

Coffee is the one of the most common beverages worldwide and has received considerable attention for its beneficial health effects. However, the association of coffee with hearing and tinnitus has not been well studied. The aim of this study was to investigate the association of coffee with hearing and tinnitus based on a national population-based survey. We evaluated hearing and tinnitus data from the 2009–2012 Korean National Health and Nutrition Examination Survey and their relationship with a coffee consumption survey. All patients underwent a medical interview, physical examination, hearing test, tinnitus questionnaire and nutrition examination. Multivariable logistic regression models were used to examine the associations between coffee and hearing loss or tinnitus. We evaluated 13,448 participants (≥19 years) participants. The frequency of coffee consumption had a statistically significant inverse correlation with bilateral hearing loss in the 40–64 years age group. Daily coffee consumers had 50–70% less hearing loss than rare coffee consumers, which tended to be a dose-dependent relationship. In addition, the frequency of coffee consumption had an inverse correlation with tinnitus in the 19–64 years age group but its association was related with hearing. Brewed coffee had more of an association than instant or canned coffee in the 40–64 years age group. These results suggest a protective effect of coffee on hearing loss and tinnitus.