Deep learning detects and visualizes bleeding events in electronic health records.

BACKGROUND: Bleeding is associated with a significantly increased morbidity and mortality. Bleeding events are often described in the unstructured text of electronic health records, which makes them difficult to identify by manual inspection. OBJECTIVES: To develop a deep learning model that detects and visualizes bleeding events in electronic health records. PATIENTS/METHODS: Three hundred electronic health records with International Classification of Diseases, Tenth Revision diagnosis codes for bleeding or leukemia were extracted. Each sentence in the electronic health record was annotated as positive or negative for bleeding. The annotated sentences were used to develop a deep learning model that detects bleeding at sentence and note level. RESULTS: On a balanced test set of 1178   sentences, the best-performing deep learning model achieved a sensitivity of 0.90, specificity of 0.90, and negative predictive value of 0.90. On a test set consisting of 700 notes, of which 49 were positive for bleeding, the model achieved a note-level sensitivity of 1.00, specificity of 0.52, and negative predictive value of 1.00. By using a sentence-level model on a note level, the model can explain its predictions by visualizing the exact sentence in a note that contains information regarding bleeding. Moreover, we found that the model performed consistently well across different types of bleedings. CONCLUSIONS: A deep learning model can be used to detect and visualize bleeding events in the free text of electronic health records. The deep learning model can thus facilitate systematic assessment of bleeding risk, and thereby optimize patient care and safety.

[Intravitreal angiostatic treatment of ocular vascular diseases].

Intraocular vascular diseases like neovascular age-related macular degeneration, diabetic macular oedema and retinal vein occlusion are associated with decline in vision if left untreated. Anti-vascular endothelia growth factor (VEGF) agents originally developed for cancer treatment were the beginning of a new treatment regime for intraocular vascular diseases. Ambiguous results exist regarding adverse effects but are considered limited. In the future, anti-VEGF agents are likely to be used for treatment of proliferative diabetic retinopathy, neovascular glaucoma and premature retinopathy.

Chronic pain patients with possible co-morbid post-traumatic stress disorder admitted to multidisciplinary pain rehabilitation-a 1-year cohort study.

BACKGROUND: Although post-traumatic stress disorder (PTSD) is a common co-morbidity in chronic pain, little is known about the association between PTSD and pain in the context of chronic pain rehabilitation. OBJECTIVE: The aim of the present study was two-fold: (1) to investigate the association of a possible PTSD diagnosis with symptoms of pain, physical and mental functioning, as well as the use of opioids, and (2) to compare the outcome of multidisciplinary chronic pain rehabilitation for patients with a possible PTSD diagnosis at admission with patients without PTSD at admission. METHOD: A consecutively referred cohort of 194 patients completed a baseline questionnaire at admission covering post-traumatic stress, pain symptoms, physical and mental functioning, as well as self-reported sleep quality and cognitive difficulties. Medication use was calculated from their medical records. A total of 95 were admitted to further multidisciplinary treatment and included in the outcome study. RESULTS: A high prevalence of possible PTSD was found (26.3%). Patients with possible co-morbid PTSD experienced significantly poorer general and mental health, poorer sleep quality, and more cognitive problems as well as inferior social functioning compared to patients without PTSD. Possible co-morbid PTSD did not result in higher use of opioids or sedatives. Surprisingly, possible co-morbid PTSD at admission was not associated with lower levels of symptom reduction from pre- to post-treatment. CONCLUSIONS: Possible co-morbid PTSD in chronic pain is a major problem associated with significantly poorer functioning on several domains. Nevertheless, our results indicate that pain-related symptoms could be treated with success despite possible co-morbid PTSD. However, since PTSD was only measured at admission it is not known whether rehabilitation actually reduced PTSD.