Automated System to Capture Patient Symptoms From Multitype Japanese Clinical Texts: Retrospective Study.

BACKGROUND: Natural language processing (NLP) techniques can be used to analyze large amounts of electronic health record texts, which encompasses various types of patient information such as quality of life, effectiveness of treatments, and adverse drug event (ADE) signals. As different aspects of a patient's status are stored in different types of documents, we propose an NLP system capable of processing 6 types of documents: physician progress notes, discharge summaries, radiology reports, radioisotope reports, nursing records, and pharmacist progress notes. OBJECTIVE: This study aimed to investigate the system's performance in detecting ADEs by evaluating the results from multitype texts. The main objective is to detect adverse events accurately using an NLP system. METHODS: We used data written in Japanese from 2289 patients with breast cancer, including medication data, physician progress notes, discharge summaries, radiology reports, radioisotope reports, nursing records, and pharmacist progress notes. Our system performs 3 processes: named entity recognition, normalization of symptoms, and aggregation of multiple types of documents from multiple patients. Among all patients with breast cancer, 103 and 112 with peripheral neuropathy (PN) received paclitaxel or docetaxel, respectively. We evaluate the utility of using multiple types of documents by correlation coefficient and regression analysis to compare their performance with each single type of document. All evaluations of detection rates with our system are performed 30 days after drug administration. RESULTS: Our system underestimates by 13.3 percentage points (74.0%-60.7%), as the incidence of paclitaxel-induced PN was 60.7%, compared with 74.0% in the previous research based on manual extraction. The Pearson correlation coefficient between the manual extraction and system results was 0.87 Although the pharmacist progress notes had the highest detection rate among each type of document, the rate did not match the performance using all documents. The estimated median duration of PN with paclitaxel was 92 days, whereas the previously reported median duration of PN with paclitaxel was 727 days. The number of events detected in each document was highest in the physician's progress notes, followed by the pharmacist's and nursing records. CONCLUSIONS: Considering the inherent cost that requires constant monitoring of the patient's condition, such as the treatment of PN, our system has a significant advantage in that it can immediately estimate the treatment duration without fine-tuning a new NLP model. Leveraging multitype documents is better than using single-type documents to improve detection performance. Although the onset time estimation was relatively accurate, the duration might have been influenced by the length of the data follow-up period. The results suggest that our method using various types of data can detect more ADEs from clinical documents.

Climate change and shifts in spring phenology of three horticultural woody perennials in northeastern USA.

We evaluated spring phenology changes from 1965 to 2001 in northeastern USA utilizing a unique data set from 72 locations with genetically identical lilac plants (Syringa chinensis, clone "Red Rothomagensis"). We also utilized a previously validated lilac-honeysuckle "spring index" model to reconstruct a more complete record of first leaf date (FLD) and first flower date (FFD) for the region from historical weather data. In addition, we examined mid-bloom dates for apple (Malus domestica) and grape (Vitis vinifera) collected at several sites in the region during approximately the same time period. Almost all lilac sites with significant linear trends for FLD or FFD versus year had negative slopes (advanced development). Regression analysis of pooled data for the 72 sites indicated an advance of -0.092 day/year for FFD (P=0.003). The slope for FLD was also negative (-0.048 day/year), but not significant (P=0.234). The simulated data from the "spring index" model, which relies on local daily temperature records, indicated highly significant (P<0.001) negative slopes of -0.210 and -0.123 day/year for FLD and FFD, respectively. Data collected for apple and grape also indicated advance spring development, with slopes for mid-bloom date versus year of -0.20 day/year (P=0.01) and -0.146 (P=0.14), respectively. Collectively, these results indicate an advance in spring phenology ranging from 2 to 8 days for these woody perennials in northeastern USA for the period 1965 to 2001, qualitatively consistent with a warming trend, and consistent with phenology shifts reported for other mid- and high-latitude regions.

Assessing vulnerability to invasion by nonnative plant species at multiple spatial scales.

Basic information on where nonnative plant species have successfully invaded is lacking. We assessed the vulnerability of 22 vegetation types (25 sets of four plots in nine study areas) to nonnative plant invasions in the north-central United States. In general, habitats with high native species richness were more heavily invaded than species-poor habitats, low-elevation areas were more invaded than high-elevation areas, and riparian zones were more invaded than nearby upland sites. For the 100 1000-m2 plots (across all vegetation types), 50% of the variation in nonnative species richness was explained by longitude, latitude, native plant species richness, soil total percentage nitrogen, and mean maximum July temperature (n = 100 plots; P < 0.001). At the vegetation-type scale (n = 25 sets of four 1000-m2 plots/type), 64% of the variation in nonnative species richness was explained by native plant species richness, elevation, and October to June precipitation (P < 0.001). The foliar cover of nonnative species (log) was strongly positively correlated with the nonnative species richness at the plot scale (r = 0.77, P < 0.001) and vegetation-type scale (r = 0.83, P < 0.001). We concluded that, at the vegetation-type and regional scales in the north-central United States, (1) vegetation types rich in native species are often highly vulnerable to invasion by nonnative plant species; (2) where several nonnative species become established, nonnative species cover can substantially increase; (3) the attributes that maintain high native plant species richness (high light, water, nitrogen, and temperatures) also help maintain nonnative plant species richness; and (4) more care must be taken to preserve native species diversity in highly vulnerable habitats.