Use of a support vector machine for categorizing free-text notes: assessment of accuracy across two institutions

Background

Electronic health record (EHR) users must regularly review large amounts of data in order to make informed clinical decisions, and such review is time-consuming and often overwhelming. Technologies like automated summarization tools, EHR search engines and natural language processing have been shown to help clinicians manage this information.

Objective

To develop a support vector machine (SVM)-based system for identifying EHR progress notes pertaining to diabetes, and to validate it at two institutions.

Materials and methods

We retrieved 2000 EHR progress notes from patients with diabetes at the Brigham and Women''s Hospital (1000 for training and 1000 for testing) and another 1000 notes from the University of Texas Physicians (for validation). We manually annotated all notes and trained a SVM using a bag of words approach. We then used the SVM on the testing and validation sets and evaluated its performance with the area under the curve (AUC) and F statistics.

Results

The model accurately identified diabetes-related notes in both the Brigham and Women''s Hospital testing set (AUC=0.956, F=0.934) and the external University of Texas Faculty Physicians validation set (AUC=0.947, F=0.935).

Discussion

Overall, the model we developed was quite accurate. Furthermore, it generalized, without loss of accuracy, to another institution with a different EHR and a distinct patient and provider population.

Conclusions

It is possible to use a SVM-based classifier to identify EHR progress notes pertaining to diabetes, and the model generalizes well.     

Integrating existing natural language processing tools for medication extraction from discharge summaries

Objective

To develop an automated system to extract medications and related information from discharge summaries as part of the 2009 i2b2 natural language processing (NLP) challenge. This task required accurate recognition of medication name, dosage, mode, frequency, duration, and reason for drug administration.

Design

We developed an integrated system using several existing NLP components developed at Vanderbilt University Medical Center, which included MedEx (to extract medication information), SecTag (a section identification system for clinical notes), a sentence splitter, and a spell checker for drug names. Our goal was to achieve good performance with minimal to no specific training for this document corpus; thus, evaluating the portability of those NLP tools beyond their home institution. The integrated system was developed using 17 notes that were annotated by the organizers and evaluated using 251 notes that were annotated by participating teams.

Measurements

The i2b2 challenge used standard measures, including precision, recall, and F-measure, to evaluate the performance of participating systems. There were two ways to determine whether an extracted textual finding is correct or not: exact matching or inexact matching. The overall performance for all six types of medication-related findings across 251 annotated notes was considered as the primary metric in the challenge.

Results

Our system achieved an overall F-measure of 0.821 for exact matching (0.839 precision; 0.803 recall) and 0.822 for inexact matching (0.866 precision; 0.782 recall). The system ranked second out of 20 participating teams on overall performance at extracting medications and related information.

Conclusions

The results show that the existing MedEx system, together with other NLP components, can extract medication information in clinical text from institutions other than the site of algorithm development with reasonable performance.     

Quantifying clinical narrative redundancy in an electronic health record

Objective

Although electronic notes have advantages compared to handwritten notes, they take longer to write and promote information redundancy in electronic health records (EHRs). We sought to quantify redundancy in clinical documentation by studying collections of physician notes in an EHR.

Design and methods

We implemented a retrospective design to gather all electronic admission, progress, resident signout and discharge summary notes written during 100 randomly selected patient admissions within a 6 month period. We modified and applied a Levenshtein edit-distance algorithm to align and compare the documents written for each of the 100 admissions. We then identified and measured the amount of text duplicated from previous notes. Finally, we manually reviewed the content that was conserved between note types in a subsample of notes.

Measurements

We measured the amount of new information in a document, which was calculated as the number of words that did not match with previous documents divided by the length, in words, of the document. Results are reported as the percentage of information in a document that had been duplicated from previously written documents.

Results

Signout and progress notes proved to be particularly redundant, with an average of 78% and 54% information duplicated from previous documents respectively. There was also significant information duplication between document types (eg, from an admission note to a progress note).

Conclusion

The study established the feasibility of exploring redundancy in the narrative record with a known sequence alignment algorithm used frequently in the field of bioinformatics. The findings provide a foundation for studying the usefulness and risks of redundancy in the EHR.     

Longitudinal analysis of pain in patients with metastatic prostate cancer using natural language processing of medical record text

Objectives

To test the feasibility of using text mining to depict meaningfully the experience of pain in patients with metastatic prostate cancer, to identify novel pain phenotypes, and to propose methods for longitudinal visualization of pain status.

Materials and methods

Text from 4409 clinical encounters for 33 men enrolled in a 15-year longitudinal clinical/molecular autopsy study of metastatic prostate cancer (Project to ELIminate lethal CANcer) was subjected to natural language processing (NLP) using Unified Medical Language System-based terms. A four-tiered pain scale was developed, and logistic regression analysis identified factors that correlated with experience of severe pain during each month.

Results

NLP identified 6387 pain and 13 827 drug mentions in the text. Graphical displays revealed the pain ‘landscape’ described in the textual records and confirmed dramatically increasing levels of pain in the last years of life in all but two patients, all of whom died from metastatic cancer. Severe pain was associated with receipt of opioids (OR=6.6, p<0.0001) and palliative radiation (OR=3.4, p=0.0002). Surprisingly, no severe or controlled pain was detected in two of 33 subjects’ clinical records. Additionally, the NLP algorithm proved generalizable in an evaluation using a separate data source (889 Informatics for Integrating Biology and the Bedside (i2b2) discharge summaries).

Discussion

Patterns in the pain experience, undetectable without the use of NLP to mine the longitudinal clinical record, were consistent with clinical expectations, suggesting that meaningful NLP-based pain status monitoring is feasible. Findings in this initial cohort suggest that ‘outlier’ pain phenotypes useful for probing the molecular basis of cancer pain may exist.

Limitations

The results are limited by a small cohort size and use of proprietary NLP software.

Conclusions

We have established the feasibility of tracking longitudinal patterns of pain by text mining of free text clinical records. These methods may be useful for monitoring pain management and identifying novel cancer phenotypes.     

Research and applications: Learning regular expressions for clinical text classification

Objectives

Natural language processing (NLP) applications typically use regular expressions that have been developed manually by human experts. Our goal is to automate both the creation and utilization of regular expressions in text classification.

Methods

We designed a novel regular expression discovery (RED) algorithm and implemented two text classifiers based on RED. The RED+ALIGN classifier combines RED with an alignment algorithm, and RED+SVM combines RED with a support vector machine (SVM) classifier. Two clinical datasets were used for testing and evaluation: the SMOKE dataset, containing 1091 text snippets describing smoking status; and the PAIN dataset, containing 702 snippets describing pain status. We performed 10-fold cross-validation to calculate accuracy, precision, recall, and F-measure metrics. In the evaluation, an SVM classifier was trained as the control.

Results

The two RED classifiers achieved 80.9–83.0% in overall accuracy on the two datasets, which is 1.3–3% higher than SVM''s accuracy (p<0.001). Similarly, small but consistent improvements have been observed in precision, recall, and F-measure when RED classifiers are compared with SVM alone. More significantly, RED+ALIGN correctly classified many instances that were misclassified by the SVM classifier (8.1–10.3% of the total instances and 43.8–53.0% of SVM''s misclassifications).

Conclusions

Machine-generated regular expressions can be effectively used in clinical text classification. The regular expression-based classifier can be combined with other classifiers, like SVM, to improve classification performance.     

Research and applications: N-gram support vector machines for scalable procedure and diagnosis classification,with applications to clinical free text data from the intensive care unit

Background

Existing risk adjustment models for intensive care unit (ICU) outcomes rely on manual abstraction of patient-level predictors from medical charts. Developing an automated method for abstracting these data from free text might reduce cost and data collection times.

Objective

To develop a support vector machine (SVM) classifier capable of identifying a range of procedures and diagnoses in ICU clinical notes for use in risk adjustment.

Materials and methods

We selected notes from 2001–2008 for 4191 neonatal ICU (NICU) and 2198 adult ICU patients from the MIMIC-II database from the Beth Israel Deaconess Medical Center. Using these notes, we developed an implementation of the SVM classifier to identify procedures (mechanical ventilation and phototherapy in NICU notes) and diagnoses (jaundice in NICU and intracranial hemorrhage (ICH) in adult ICU). On the jaundice classification task, we also compared classifier performance using n-gram features to unigrams with application of a negation algorithm (NegEx).

Results

Our classifier accurately identified mechanical ventilation (accuracy=0.982, F₁=0.954) and phototherapy use (accuracy=0.940, F₁=0.912), as well as jaundice (accuracy=0.898, F₁=0.884) and ICH diagnoses (accuracy=0.938, F₁=0.943). Including bigram features improved performance on the jaundice (accuracy=0.898 vs 0.865) and ICH (0.938 vs 0.927) tasks, and outperformed NegEx-derived unigram features (accuracy=0.898 vs 0.863) on the jaundice task.

Discussion

Overall, a classifier using n-gram support vectors displayed excellent performance characteristics. The classifier generalizes to diverse patient populations, diagnoses, and procedures.

Conclusions

SVM-based classifiers can accurately identify procedure status and diagnoses among ICU patients, and including n-gram features improves performance, compared to existing methods.     

Research and applications: A comprehensive study of named entity recognition in Chinese clinical text

Objective

Named entity recognition (NER) is one of the fundamental tasks in natural language processing. In the medical domain, there have been a number of studies on NER in English clinical notes; however, very limited NER research has been carried out on clinical notes written in Chinese. The goal of this study was to systematically investigate features and machine learning algorithms for NER in Chinese clinical text.

Materials and methods

We randomly selected 400 admission notes and 400 discharge summaries from Peking Union Medical College Hospital in China. For each note, four types of entity—clinical problems, procedures, laboratory test, and medications—were annotated according to a predefined guideline. Two-thirds of the 400 notes were used to train the NER systems and one-third for testing. We investigated the effects of different types of feature including bag-of-characters, word segmentation, part-of-speech, and section information, and different machine learning algorithms including conditional random fields (CRF), support vector machines (SVM), maximum entropy (ME), and structural SVM (SSVM) on the Chinese clinical NER task. All classifiers were trained on the training dataset and evaluated on the test set, and micro-averaged precision, recall, and F-measure were reported.

Results

Our evaluation on the independent test set showed that most types of feature were beneficial to Chinese NER systems, although the improvements were limited. The system achieved the highest performance by combining word segmentation and section information, indicating that these two types of feature complement each other. When the same types of optimized feature were used, CRF and SSVM outperformed SVM and ME. More specifically, SSVM achieved the highest performance of the four algorithms, with F-measures of 93.51% and 90.01% for admission notes and discharge summaries, respectively.     

à la Recherche du Temps Perdu: extracting temporal relations from medical text in the 2012 i2b2 NLP challenge

Objective

An analysis of the timing of events is critical for a deeper understanding of the course of events within a patient record. The 2012 i2b2 NLP challenge focused on the extraction of temporal relationships between concepts within textual hospital discharge summaries.

Materials and methods

The team from the National Research Council Canada (NRC) submitted three system runs to the second track of the challenge: typifying the time-relationship between pre-annotated entities. The NRC system was designed around four specialist modules containing statistical machine learning classifiers. Each specialist targeted distinct sets of relationships: local relationships, ‘sectime’-type relationships, non-local overlap-type relationships, and non-local causal relationships.

Results

The best NRC submission achieved a precision of 0.7499, a recall of 0.6431, and an F1 score of 0.6924, resulting in a statistical tie for first place. Post hoc improvements led to a precision of 0.7537, a recall of 0.6455, and an F1 score of 0.6954, giving the highest scores reported on this task to date.

Discussion and conclusions

Methods for general relation extraction extended well to temporal relations, and gave top-ranked state-of-the-art results. Careful ordering of predictions within result sets proved critical to this success.     

Retrorectal tumours: literature review and Vilnius University Hospital "Santariskiu klinikos" experience of 14 cases

Objective

Retrorectal tumours are rare lesions in adults. The diagnosis of retrorectal lesion is often difficult and misdiagnosis is common. We present significant number of cases in view of scarce information available on this matter.

Methods

14 patients were treated at Vilnius university hospital "Santariskiu klinikos" Centre of abdominal surgery from 1997 to 2010. The case notes of patients who underwent surgery for a retrorectal tumour were reviewed retrospectively. Surgical histories, operations, histological tumour type, surgical time, weight of the specimen, blood loss, length of stay were analysed.

Results

13 patients underwent laparotomy, 1 patient had combined perineal approach and laparotomy. The most common types of the tumour were fibroma (3 cases), leiomyosarcoma (2 cases). 5 tumours (35,7%) were found to be malignant. 57% of the patients had undergone at least one operation prior to definitive treatment. 5 female patients were initially admitted under gynaecologists. Hospital stay varied from 14 days to 22 days (mean 16,2 days). A report of a representative case is presented.

Conclusions

Retrorectal lesions in female patients can mimic gynaecological pathology. Patients with this rare pathology are to be treated in a major tertiary hospital by surgeons, who are able to operate safely in the retrorectal space.     

Automated concept-level information extraction to reduce the need for custom software and rules development

Objective

Despite at least 40 years of promising empirical performance, very few clinical natural language processing (NLP) or information extraction systems currently contribute to medical science or care. The authors address this gap by reducing the need for custom software and rules development with a graphical user interface-driven, highly generalizable approach to concept-level retrieval.

Materials and methods

A ‘learn by example’ approach combines features derived from open-source NLP pipelines with open-source machine learning classifiers to automatically and iteratively evaluate top-performing configurations. The Fourth i2b2/VA Shared Task Challenge''s concept extraction task provided the data sets and metrics used to evaluate performance.

Results

Top F-measure scores for each of the tasks were medical problems (0.83), treatments (0.82), and tests (0.83). Recall lagged precision in all experiments. Precision was near or above 0.90 in all tasks.

Discussion

With no customization for the tasks and less than 5 min of end-user time to configure and launch each experiment, the average F-measure was 0.83, one point behind the mean F-measure of the 22 entrants in the competition. Strong precision scores indicate the potential of applying the approach for more specific clinical information extraction tasks. There was not one best configuration, supporting an iterative approach to model creation.

Conclusion

Acceptable levels of performance can be achieved using fully automated and generalizable approaches to concept-level information extraction. The described implementation and related documentation is available for download.     

Research and applications: Patient-level temporal aggregation for text-based asthma status ascertainment

Objective

To specify the problem of patient-level temporal aggregation from clinical text and introduce several probabilistic methods for addressing that problem. The patient-level perspective differs from the prevailing natural language processing (NLP) practice of evaluating at the term, event, sentence, document, or visit level.

Methods

We utilized an existing pediatric asthma cohort with manual annotations. After generating a basic feature set via standard clinical NLP methods, we introduce six methods of aggregating time-distributed features from the document level to the patient level. These aggregation methods are used to classify patients according to their asthma status in two hypothetical settings: retrospective epidemiology and clinical decision support.

Results

In both settings, solid patient classification performance was obtained with machine learning algorithms on a number of evidence aggregation methods, with Sum aggregation obtaining the highest F₁ score of 85.71% on the retrospective epidemiological setting, and a probability density function-based method obtaining the highest F₁ score of 74.63% on the clinical decision support setting. Multiple techniques also estimated the diagnosis date (index date) of asthma with promising accuracy.

Discussion

The clinical decision support setting is a more difficult problem. We rule out some aggregation methods rather than determining the best overall aggregation method, since our preliminary data set represented a practical setting in which manually annotated data were limited.

Conclusion

Results contrasted the strengths of several aggregation algorithms in different settings. Multiple approaches exhibited good patient classification performance, and also predicted the timing of estimates with reasonable accuracy.     

Evaluating the utility of syndromic surveillance algorithms for screening to detect potentially clonal hospital infection outbreaks

Objective

The authors evaluated algorithms commonly used in syndromic surveillance for use as screening tools to detect potentially clonal outbreaks for review by infection control practitioners.

Design

Study phase 1 applied four aberrancy detection algorithms (CUSUM, EWMA, space-time scan statistic, and WSARE) to retrospective microbiologic culture data, producing a list of past candidate outbreak clusters. In phase 2, four infectious disease physicians categorized the phase 1 algorithm-identified clusters to ascertain algorithm performance. In phase 3, project members combined the algorithms to create a unified screening system and conducted a retrospective pilot evaluation.

Measurements

The study calculated recall and precision for each algorithm, and created precision-recall curves for various methods of combining the algorithms into a unified screening tool.

Results

Individual algorithm recall and precision ranged from 0.21 to 0.31 and from 0.053 to 0.29, respectively. Few candidate outbreak clusters were identified by more than one algorithm. The best method of combining the algorithms yielded an area under the precision-recall curve of 0.553. The phase 3 combined system detected all infection control-confirmed outbreaks during the retrospective evaluation period.

Limitations

Lack of phase 2 reviewers'' agreement indicates that subjective expert review was an imperfect gold standard. Less conservative filtering of culture results and alternate parameter selection for each algorithm might have improved algorithm performance.

Conclusion

Hospital outbreak detection presents different challenges than traditional syndromic surveillance. Nevertheless, algorithms developed for syndromic surveillance have potential to form the basis of a combined system that might perform clinically useful hospital outbreak screening.     

Research and applications: Influenza detection from emergency department reports using natural language processing and Bayesian network classifiers

Objectives

To evaluate factors affecting performance of influenza detection, including accuracy of natural language processing (NLP), discriminative ability of Bayesian network (BN) classifiers, and feature selection.

Methods

We derived a testing dataset of 124 influenza patients and 87 non-influenza (shigellosis) patients. To assess NLP finding-extraction performance, we measured the overall accuracy, recall, and precision of Topaz and MedLEE parsers for 31 influenza-related findings against a reference standard established by three physician reviewers. To elucidate the relative contribution of NLP and BN classifier to classification performance, we compared the discriminative ability of nine combinations of finding-extraction methods (expert, Topaz, and MedLEE) and classifiers (one human-parameterized BN and two machine-parameterized BNs). To assess the effects of feature selection, we conducted secondary analyses of discriminative ability using the most influential findings defined by their likelihood ratios.

Results

The overall accuracy of Topaz was significantly better than MedLEE (with post-processing) (0.78 vs 0.71, p<0.0001). Classifiers using human-annotated findings were superior to classifiers using Topaz/MedLEE-extracted findings (average area under the receiver operating characteristic (AUROC): 0.75 vs 0.68, p=0.0113), and machine-parameterized classifiers were superior to the human-parameterized classifier (average AUROC: 0.73 vs 0.66, p=0.0059). The classifiers using the 17 ‘most influential’ findings were more accurate than classifiers using all 31 subject-matter expert-identified findings (average AUROC: 0.76>0.70, p<0.05).

Conclusions

Using a three-component evaluation method we demonstrated how one could elucidate the relative contributions of components under an integrated framework. To improve classification performance, this study encourages researchers to improve NLP accuracy, use a machine-parameterized classifier, and apply feature selection methods.     

A sequence labeling approach to link medications and their attributes in clinical notes and clinical trial announcements for information extraction

Objective

The goal of this work was to evaluate machine learning methods, binary classification and sequence labeling, for medication–attribute linkage detection in two clinical corpora.

Data and methods

We double annotated 3000 clinical trial announcements (CTA) and 1655 clinical notes (CN) for medication named entities and their attributes. A binary support vector machine (SVM) classification method with parsimonious feature sets, and a conditional random fields (CRF)-based multi-layered sequence labeling (MLSL) model were proposed to identify the linkages between the entities and their corresponding attributes. We evaluated the system''s performance against the human-generated gold standard.

Results

The experiments showed that the two machine learning approaches performed statistically significantly better than the baseline rule-based approach. The binary SVM classification achieved 0.94 F-measure with individual tokens as features. The SVM model trained on a parsimonious feature set achieved 0.81 F-measure for CN and 0.87 for CTA. The CRF MLSL method achieved 0.80 F-measure on both corpora.

Discussion and conclusions

We compared the novel MLSL method with a binary classification and a rule-based method. The MLSL method performed statistically significantly better than the rule-based method. However, the SVM-based binary classification method was statistically significantly better than the MLSL method for both the CTA and CN corpora. Using parsimonious feature sets both the SVM-based binary classification and CRF-based MLSL methods achieved high performance in detecting medication name and attribute linkages in CTA and CN.     

Using rule-based natural language processing to improve disease normalization in biomedical text

Background and objective

In order for computers to extract useful information from unstructured text, a concept normalization system is needed to link relevant concepts in a text to sources that contain further information about the concept. Popular concept normalization tools in the biomedical field are dictionary-based. In this study we investigate the usefulness of natural language processing (NLP) as an adjunct to dictionary-based concept normalization.

Methods

We compared the performance of two biomedical concept normalization systems, MetaMap and Peregrine, on the Arizona Disease Corpus, with and without the use of a rule-based NLP module. Performance was assessed for exact and inexact boundary matching of the system annotations with those of the gold standard and for concept identifier matching.

Results

Without the NLP module, MetaMap and Peregrine attained F-scores of 61.0% and 63.9%, respectively, for exact boundary matching, and 55.1% and 56.9% for concept identifier matching. With the aid of the NLP module, the F-scores of MetaMap and Peregrine improved to 73.3% and 78.0% for boundary matching, and to 66.2% and 69.8% for concept identifier matching. For inexact boundary matching, performances further increased to 85.5% and 85.4%, and to 73.6% and 73.3% for concept identifier matching.

Conclusions

We have shown the added value of NLP for the recognition and normalization of diseases with MetaMap and Peregrine. The NLP module is general and can be applied in combination with any concept normalization system. Whether its use for concept types other than disease is equally advantageous remains to be investigated.     

Biomedical negation scope detection with conditional random fields

Objective

Negation is a linguistic phenomenon that marks the absence of an entity or event. Negated events are frequently reported in both biological literature and clinical notes. Text mining applications benefit from the detection of negation and its scope. However, due to the complexity of language, identifying the scope of negation in a sentence is not a trivial task.

Design

Conditional random fields (CRF), a supervised machine-learning algorithm, were used to train models to detect negation cue phrases and their scope in both biological literature and clinical notes. The models were trained on the publicly available BioScope corpus.

Measurement

The performance of the CRF models was evaluated on identifying the negation cue phrases and their scope by calculating recall, precision and F1-score. The models were compared with four competitive baseline systems.

Results

The best CRF-based model performed statistically better than all baseline systems and NegEx, achieving an F1-score of 98% and 95% on detecting negation cue phrases and their scope in clinical notes, and an F1-score of 97% and 85% on detecting negation cue phrases and their scope in biological literature.

Conclusions

This approach is robust, as it can identify negation scope in both biological and clinical text. To benefit text mining applications, the system is publicly available as a Java API and as an online application at http://negscope.askhermes.org.     

Use of electronic clinical documentation: time spent and team interactions

Objective

To measure the time spent authoring and viewing documentation and to study patterns of usage in healthcare practice.

Design

Audit logs for an electronic health record were used to calculate rates, and social network analysis was applied to ascertain usage patterns. Subjects comprised all care providers at an urban academic medical center who authored or viewed electronic documentation.

Measurement

Rate and time of authoring and viewing clinical documentation, and associations among users were measured.

Results

Users spent 20–103 min per day authoring notes and 7–56 min per day viewing notes, with physicians spending less than 90 min per day total. About 16% of attendings'' notes, 8% of residents'' notes, and 38% of nurses'' notes went unread by other users, and, overall, 16% of notes were never read by anyone. Viewing of notes dropped quickly with the age of the note, but notes were read at a low but measurable rate, even after 2 years. Most healthcare teams (77%) included a nurse, an attending, and a resident, and those three users'' groups were the first to write notes during an admission.

Limitations

The limitations were restriction to a single academic medical center and use of log files without direct observation.

Conclusions

Care providers spend a significant amount of time viewing and authoring notes. Many notes are never read, and rates of usage vary significantly by author and viewer. While the rate of viewing a note drops quickly with its age, even after 2 years inpatient notes are still viewed.     

Towards comprehensive syntactic and semantic annotations of the clinical narrative

Objective

To create annotated clinical narratives with layers of syntactic and semantic labels to facilitate advances in clinical natural language processing (NLP). To develop NLP algorithms and open source components.

Methods

Manual annotation of a clinical narrative corpus of 127 606 tokens following the Treebank schema for syntactic information, PropBank schema for predicate-argument structures, and the Unified Medical Language System (UMLS) schema for semantic information. NLP components were developed.

Results

The final corpus consists of 13 091 sentences containing 1772 distinct predicate lemmas. Of the 766 newly created PropBank frames, 74 are verbs. There are 28 539 named entity (NE) annotations spread over 15 UMLS semantic groups, one UMLS semantic type, and the Person semantic category. The most frequent annotations belong to the UMLS semantic groups of Procedures (15.71%), Disorders (14.74%), Concepts and Ideas (15.10%), Anatomy (12.80%), Chemicals and Drugs (7.49%), and the UMLS semantic type of Sign or Symptom (12.46%). Inter-annotator agreement results: Treebank (0.926), PropBank (0.891–0.931), NE (0.697–0.750). The part-of-speech tagger, constituency parser, dependency parser, and semantic role labeler are built from the corpus and released open source. A significant limitation uncovered by this project is the need for the NLP community to develop a widely agreed-upon schema for the annotation of clinical concepts and their relations.

Conclusions

This project takes a foundational step towards bringing the field of clinical NLP up to par with NLP in the general domain. The corpus creation and NLP components provide a resource for research and application development that would have been previously impossible.     

Perinatal mortality audit at Tikure Anbessa Teaching Hospital,Addis Ababa,Ethiopia: 1995 to 1996

Objectives

To determine the prevalence and causes of perinatal mortality rates at Tikure Anbessa hospital, Ethiopia, 1995–96

Design

A cross-sectional review of hospital records of all women who delivered at Tikure Anbessa Hospital. Data were collected prospectively.

Setting

Teaching Hospital of Tikure Anbessa, Addis Ababa, Ethiopia

Subjects

Women and neonates from 8986 deliveries. Deliveries exceeding 28 weeks of gestation or birth weight of 1000 grams were considered.

Materials and Methods

All deliveries were included for infants whose gestational age exceeded 28 completed weeks. When the gestational age was unknown, the birth weight of 1000 grams or more was considered. All perinatal deaths and obstetric complications were identified. Labour chart, mode of delivery and summary of delivery had been recorded by the responsible resident at a monthly combined obstetric and paediatric perinatal mortality meeting. Age of the mother, parity, booking status for antenatal care, obstetric complications, labour, mode of delivery, birth weight, gestational age, one and five minutes Apgar scores were collected.

Results

A total of 8986 deliveries were conducted during the study period. Of these 6933 (77.2%) were booked for antenatal care while the remaining 2053 (22.8%) were not booked in any health institution. The perinatal mortality rate of the hospital was 71.6 per 1000 live births. The risk of perinatal mortality was more than doubled among mothers who failed to book for antenatal care follow-up and no laboratory investigations was done were birth asphyxia followed by premature birth 15.4% and 12% respectively.

Conclusions

Perinatal mortality rates are high at Tikure Anbessa Hospital, Ethiopia. There is need to ensure that pregnant women are booked for antenatal care so as to provide adequate antenatal and perinatal health care.