Metastatic Lymph Node Ratio of Central Neck Compartment Has Predictive Values for Locoregional Recurrence in Papillary Thyroid Microcarcinoma

ObjectivesThis study aimed to evaluate the significance of metastatic lymph node ratio (the ratio between the metastatic lymph node and the harvested lymph nodes; MLNR) in the central neck for the prediction of locoregional recurrence in patients with papillary thyroid microcarcinoma.MethodsAfter reviewing medical records of papillary thyroid microcarcinoma patients who received total thyroidectomy with central neck node dissection, 573 consecutive adult patients were enrolled in this study, with a follow-up period of more than 36 months. Regarding the risk of recurrence, multivariate analyses were performed with the following variables; sex, age, multiplicity of the primary tumor, presence of pathological extrathyroidal extension, the level of postoperative stimulated serum thyroglobulin, the number of harvested lymph nodes, the number of lymph node metastasis and MLNR.ResultsThe MLNR showed a predictive significance for the locoregional recurrence (P<0.05). Most recurrences were occurred in the lateral neck (n=12, 80%) with a median interval of 20 months. The lowest cutoff value of the MLNR for a meaningful separation of disease recurrence was 0.44 (hazard ratio, 8.86; 95% confidence interval, 1.49 to 52.58; P=0.001).ConclusionWhen the MLNR is higher than 0.44, there is an increased risk of locoregional recurrence mostly in the lateral neck. Therefore, MLNR of the central neck in a permanent or frozen biopsy may be helpful in decision making in the extent of thyroidectomy and/or the need for contralateral central neck lymph nodes dissection.     

Recognition of medication information from discharge summaries using ensembles of classifiers

ABSTRACT: BACKGROUND: Extraction of clinical information such as medications or problems from clinical text is an important task of clinical natural language processing (NLP). Rule-based methods are often used in clinical NLP systems because they are easy to adapt and customize. Recently, supervised machine learning methods have proven to be effective in clinical NLP as well. However, combining different classifiers to further improve the performance of clinical entity recognition systems has not been investigated extensively. Combining classifiers into an ensemble classifier presents both challenges and opportunities to improve performance in such NLP tasks. METHODS: We investigated ensemble classifiers that used different voting strategies to combine outputs from three individual classifiers: a rule-based system, a support vector machine (SVM) based system, and a conditional random field (CRF) based system. Three voting methods were proposed and evaluated using the annotated data sets from the 2009 i2b2 NLP challenge: simple majority, local SVM-based voting, and local CRF-based voting. RESULTS: Evaluation on 268 manually annotated discharge summaries from the i2b2 challenge showed that the local CRF-based voting method achieved the best F-score of 90.84% (94.11% Precision, 87.81% Recall) for 10-fold cross-validation. We then compared our systems with the first-ranked system in the challenge by using the same training and test sets. Our system based on majority voting achieved a better F-score of 89.65% (93.91% Precision, 85.76% Recall) than the previously reported F-score of 89.19% (93.78% Precision, 85.03% Recall) by the first-ranked system in the challenge. CONCLUSIONS: Our experimental results using the 2009 i2b2 challenge datasets showed that ensemble classifiers that combine individual classifiers into a voting system could achieve better performance than a single classifier in recognizing medication information from clinical text. It suggests that simple strategies that can be easily implemented such as majority voting could have the potential to significantly improve clinical entity recognition.