Discovering themes in medical records of patients with psychogenic non-epileptic seizures.

INTRODUCTION: Epileptic and psychogenic non-epileptic seizures (PNES) are common diagnostic problems encountered in hospital practice. This study explores the use of unsupervised machine learning in discovering themes in medical records of patients presenting with PNES. We hypothesised that themes generated by machine learning are comparable with the classification by human experts. METHODS: This is a retrospective analysis of the medical records in the emergency department of patients (age >18 years) with PNES who underwent inpatient video-electroencephalography monitoring from May 2009 to June 2014 and received a final diagnosis of PNES. Prior to machine learning of written text, we applied a standardised approach in natural language processing to create a document-term matrix (removal of numbers, stop-words and punctuations, transforming fonts to lower case). The words were separated into tokens and treated as if existing within a bag-of-words. A probability of each word existing within a topic (theme) was modelled on multivariate Dirichlet distribution (R Foundation, V.3.5.0). Next, we asked four experts to independently provide a clinical interpretation of the generated topics. When the majority of (≥3) experts agreed, it was regarded as highly congruent. Interactive data are available on the web at (https://gntem2.github.io/PNES/%23topic=1&lambda=0.6&term=). RESULTS: There were 39 patients (74.4% women, median age 35 years with range 20-82). A total of 121 documents were converted to text files for text mining. There were 15 generated topics with 12/15 topics rated as highly congruent. The main themes were about descriptors of seizures and medication use. CONCLUSIONS: The findings from machine learning on PNES-related documentation provides evidence for the feasibility of applying machine-learning methodology to analyse large volumes of medical records. The topics generated by machine learning were congruent with interpretations by clinicians indicating this method can be used for screening of medical conditions among large volumes of medical records.

Psychiatric comorbidity and impact on health service utilization in a community sample of patients with epilepsy.

We aimed to determine the level of psychological distress in community-treated patients with epilepsy and to determine if this distress is associated with increased health service use. The Australian National prescription database was used to recruit patients with epilepsy onto the Tasmanian Epilepsy Register (TER). Psychological distress was measured using the K10 in the TER patients and compared to the Tasmanian population using the National Health Survey 2004-5. Of the 1,180 on the TER, 43 withdrew, 36 died, and 262 were excluded. Of 839 patients, 652 completed the K10 (78%). High-very high levels of psychological distress were observed in patients with epilepsy compared with the general population [odds ratio (OR) 2.14, 95% confidence interval (CI) 1.79-2.56]. Patients with high-very high psychological distress had increased attendance at general practitioners (p < 0.001), specialists (p = 0.02), and emergency departments (p = 0.004). Psychological distress is increased in community-treated patients with epilepsy compared to the general population, and is associated with increased health service use.

Surveillance of Creutzfeldt-Jakob disease in Australia: 2008.

Australia-wide surveillance of all human transmissible spongiform encephalopathies (TSEs) is performed by the Australian National Creutzfeldt-Jakob Disease Registry (ANCJDR), since establishment in October 1993. During the surveillance period 1 April 2007 to 31 March 2008, the ANCJDR received 78 new suspect case notifications of TSEs (67 in 2007, 13 in 2008). This level of suspect case notification aligns with the previous 2006/2007 surveillance period, which was elevated in comparison to the previous 5 years. Based on the total number of probable and definite Creutzfeldt-Jakob disease (CJD) cases, encompassing retrospective cases to 1970 and prospectively ascertained cases from 1993 to 31 March 2008, the average age-adjusted mortality rate is 1.16 deaths per million per year. In this short report, we provide updated Australian TSE figures and describe recent changes in surveillance mechanisms and review their impact on case notifications and eventual CJD classification.

Creutzfeldt-Jakob disease: Australian surveillance update to March 2007.

From October 1993, prospective, national surveillance of the rare class of neurodegenerative diseases known as transmissible spongiform encephalopathies (TSEs) has been performed by the Australian National Creutzfeldt-Jakob Disease Registry. Surveillance of TSEs prior to October 1993, involved the retrospective ascertainment of TSE cases from 1970 to 1993. In this report, surveillance data for 2006 are presented in detail and compared to cumulative national TSE ascertainment as well as international experience. The higher incidence of TSEs in 2006 is not without precedent and can be attributed to higher referrals and consequent post-mortem rates.

Rippling muscle disease.

A case of rippling muscle disease is presented and features of this rare condition, and its association with caveolin-3 are discussed.