Malocclusion Classification on 3D Cone-Beam CT Craniofacial Images Using Multi-Channel Deep Learning Models.

Analyzing and interpreting cone-beam computed tomography (CBCT) images is a complicated and often time-consuming process. In this study, we present two different architectures of multi-channel deep learning (DL) models: "Ensemble" and "Synchronized multi-channel", to automatically identify and classify skeletal malocclusions from 3D CBCT craniofacial images. These multi-channel models combine three individual single-channel base models using a voting scheme and a two-step learning process, respectively, to simultaneously extract and learn a visual representation from three different directional views of 2D images generated from a single 3D CBCT image. We also employ a visualization method called "Class-selective Relevance Mapping" (CRM) to explain the learned behavior of our DL models by localizing and highlighting a discriminative area within an input image. Our multi-channel models achieve significantly better performance overall (accuracy exceeding 93%), compared to single-channel DL models that only take one specific directional view of 2D projected image as an input. In addition, CRM visually demonstrates that a DL model based on the sagittal-left view of 2D images outperforms those based on other directional 2D images.Clinical Relevance- the proposed method aims at assisting orthodontist to determine the best treatment path for the patient be it orthodontic or surgical treatment or a combination of both.

Understanding spatial language in radiology: Representation framework, annotation, and spatial relation extraction from chest X-ray reports using deep learning.

Radiology reports contain a radiologist's interpretations of images, and these images frequently describe spatial relations. Important radiographic findings are mostly described in reference to an anatomical location through spatial prepositions. Such spatial relationships are also linked to various differential diagnoses and often described through uncertainty phrases. Structured representation of this clinically significant spatial information has the potential to be used in a variety of downstream clinical informatics applications. Our focus is to extract these spatial representations from the reports. For this, we first define a representation framework based on the Spatial Role Labeling (SpRL) scheme, which we refer to as Rad-SpRL. In Rad-SpRL, common radiological entities tied to spatial relations are encoded through four spatial roles: Trajector, Landmark, Diagnosis, and Hedge, all identified in relation to a spatial preposition (or Spatial Indicator). We annotated a total of 2,000 chest X-ray reports following Rad-SpRL. We then propose a deep learning-based natural language processing (NLP) method involving word and character-level encodings to first extract the Spatial Indicators followed by identifying the corresponding spatial roles. Specifically, we use a bidirectional long short-term memory (Bi-LSTM) conditional random field (CRF) neural network as the baseline model. Additionally, we incorporate contextualized word representations from pre-trained language models (BERT and XLNet) for extracting the spatial information. We evaluate both gold and predicted Spatial Indicators to extract the four types of spatial roles. The results are promising, with the highest average F1 measure for Spatial Indicator extraction being 91.29 (XLNet); the highest average overall F1 measure considering all the four spatial roles being 92.9 using gold Indicators (XLNet); and 85.6 using predicted Indicators (BERT pre-trained on MIMIC notes). The corpus is available in Mendeley at http://dx.doi.org/10.17632/yhb26hfz8n.1 and https://github.com/krobertslab/datasets/blob/master/Rad-SpRL.xml.

Finding Understudied Disorders Potentially Associated with Maternal Morbidity and Mortality.

Objective Clinical research literature focuses primarily on the most common causes of maternal morbidity and mortality (MMM). We explore sections of the discharge summaries of pregnant or postpartum women admitted to an intensive care unit (ICU) to identify associated disorders and mine the literature to identify knowledge gaps in clinical research. Methods Data for the study were discharge summaries in the MIMIC (Medical Information Mart for Intensive Care) database. We extracted a control cohort to study if there is a difference in comorbidities between pregnant and not pregnant patients with similar reasons for admission. We identified comorbidities of the Unified Medical Language System (UMLS) semantic types disease or syndrome, Mental or behavioral dysfunction, and injury, or poisoning. We used Entrez programming utilities (E-utilities) to query PubMed ® . Results We identified 246 pregnant and postpartum patients. A control group of 587 not pregnancy related admissions matched on age and admit diagnosis. We found overlap of 24.3% discharge diagnoses between the two groups, and 7.5% of the codes exclusively in the pregnancy group. We identified 33 disease mentions not included in the most common reported causes of MMM. Conclusion Our results demonstrate that clinical text provides additional comorbidities associated with maternal complications that need further clinical research.

Semantic annotation of consumer health questions.

BACKGROUND: Consumers increasingly use online resources for their health information needs. While current search engines can address these needs to some extent, they generally do not take into account that most health information needs are complex and can only fully be expressed in natural language. Consumer health question answering (QA) systems aim to fill this gap. A major challenge in developing consumer health QA systems is extracting relevant semantic content from the natural language questions (question understanding). To develop effective question understanding tools, question corpora semantically annotated for relevant question elements are needed. In this paper, we present a two-part consumer health question corpus annotated with several semantic categories: named entities, question triggers/types, question frames, and question topic. The first part (CHQA-email) consists of relatively long email requests received by the U.S. National Library of Medicine (NLM) customer service, while the second part (CHQA-web) consists of shorter questions posed to MedlinePlus search engine as queries. Each question has been annotated by two annotators. The annotation methodology is largely the same between the two parts of the corpus; however, we also explain and justify the differences between them. Additionally, we provide information about corpus characteristics, inter-annotator agreement, and our attempts to measure annotation confidence in the absence of adjudication of annotations. RESULTS: The resulting corpus consists of 2614 questions (CHQA-email: 1740, CHQA-web: 874). Problems are the most frequent named entities, while treatment and general information questions are the most common question types. Inter-annotator agreement was generally modest: question types and topics yielded highest agreement, while the agreement for more complex frame annotations was lower. Agreement in CHQA-web was consistently higher than that in CHQA-email. Pairwise inter-annotator agreement proved most useful in estimating annotation confidence. CONCLUSIONS: To our knowledge, our corpus is the first focusing on annotation of uncurated consumer health questions. It is currently used to develop machine learning-based methods for question understanding. We make the corpus publicly available to stimulate further research on consumer health QA.

A dataset of 200 structured product labels annotated for adverse drug reactions.

Adverse drug reactions (ADRs), unintended and sometimes dangerous effects that a drug may have, are one of the leading causes of morbidity and mortality during medical care. To date, there is no structured machine-readable authoritative source of known ADRs. The United States Food and Drug Administration (FDA) partnered with the National Library of Medicine to create a pilot dataset containing standardised information about known adverse reactions for 200 FDA-approved drugs. The Structured Product Labels (SPLs), the documents FDA uses to exchange information about drugs and other products, were manually annotated for adverse reactions at the mention level to facilitate development and evaluation of text mining tools for extraction of ADRs from all SPLs. The ADRs were then normalised to the Unified Medical Language System (UMLS) and to the Medical Dictionary for Regulatory Activities (MedDRA). We present the curation process and the structure of the publicly available database SPL-ADR-200db containing 5,098 distinct ADRs. The database is available at https://bionlp.nlm.nih.gov/tac2017adversereactions/; the code for preparing and validating the data is available at https://github.com/lhncbc/fda-ars.

Finding medication doses in the liteature.

Medication doses, one of the determining factors in medication safety and effectiveness, are present in the literature, but only in free-text form. We set out to determine if the systems developed for extracting drug prescription information from clinical text would yield comparable results on scientific literature and if sequence-to-sequence learning with neural networks could improve over the current state-of-the-art. We developed a collection of 694 PubMed Central documents annotated with drug dose information using the i2b2 schema. We found that less than half of the drug doses are present in the MEDLINE/PubMed abstracts, and full-text is needed to identify the other half. We identified the differences in the scope and formatting of drug dose information in the literature and clinical text, which require developing new dose extraction approaches. Finally, we achieved 83.9% recall, 87.2% precision and 85.5% F1 score in extracting complete drug prescription information from the literature.

Mining the literature for genes associated with placenta-mediated maternal diseases.

Automated literature analysis could significantly speed up understanding of the role of the placenta and the impact of its development and functions on the health of the mother and the child. To facilitate automatic extraction of information about placenta-mediated disorders from the literature, we manually annotated genes and proteins, the associated diseases, and the functions and processes involved in the development and function of placenta in a collection of PubMed/MEDLINE abstracts. We developed three baseline approaches to finding sentences containing this information: one based on supervised machine learning (ML) and two based on distant supervision: 1) using automated detection of named entities and 2) using MeSH. We compare the performance of several well-known supervised ML algorithms and identify two approaches, Support Vector Machines (SVM) and Generalized Linear Models (GLM), which yield up to 98% recall precision and F1 score. We demonstrate that distant supervision approaches could be used at the expense of missing up to 15% of relevant documents.

Mortalidad en pacientes en edad reproductiva en la provincia de Las Tunas / Mortality in patients of childbearing age in Las Tunas province

Fundamento: las mujeres en edad fértil constituyen un grupo poblacional que tiene una gran importancia para el país, son las encargadas del aumento de la natalidad y pertenecen al conjunto de personas económicamente activas, que aportan de manera muy significativa a la sociedad.Objetivo: describir las principales causas de muerte de las mujeres en edad fértil en la provincia de Las Tunas, en el año 2013.Método: se realizó un estudio observacional, descriptivo y transversal, con el objetivo de describir las principales causas de muerte de las mujeres con edades comprendidas entre los 12 y 49 años, de la provincia de Las Tunas, en el período de enero a diciembre del año 2013. La población de estudio estuvo constituida por las 104 mujeres con estos criterios, que fallecieron durante ese período. Para la recolección de la información se utilizaron los datos estadísticos del Departamento de Estadística de la Dirección Provincial de Salud. Se procesaron los datos por el método estadístico porcentual y se calcularon tasas.Resultados: en la muestra estudiada predominó el grupo de edades comprendido entre los 35 y 49 años; los municipios de Majibacoa, Puerto Padre, Las Tunas y Jobabo fueron los más afectados, según tasas por su densidad de población. Las principales causas de muerte estuvieron representadas por los tumores malignos, las muertes por causas violentas, enfermedades del corazón y las infecciones, mientras que los tipos de cáncer más frecuentes fueron el de mama, pulmón y útero.Conclusiones: se describieron las principales causas de muerte en mujeres de Las Tunas en edad fértil, con predominio de la muerte por tumores malignos y, entre ellos, el más frecuente fue el cáncer de mama (AU)

Background: women of childbearing age are a population group that has great importance for the country; they are responsible of birth increase and belong to the group of economically active people who provides a significant contribution to society.Objective: to describe the main death causes of women of childbearing age in Las Tunas province in 2013.Method: an observational, descriptive and cross-sectional study was carried out with the objective of describing the main causes of death among women aged between 12 and 49 years in Las Tunas provincefrom January to December, 2013. The study population consisted of 104 women with these criteria who died during that period. To collect the information, the statistical data of the Statistics Department of the Provincial Health Directorate were used. Data were processed by the statistical percentage method and rates were calculated.Results: in the studied sample the age group between 35 and 49 years prevailed; Majibacoa, Puerto Padre, Las Tunas and Jobabo were the most affected municipalities, according to their population density rates. The main causes of death were malignant tumors, violent deaths, heart diseases and infections and the most frequent types of cancer were breast, lung and uterus cancer.Conclusions: the main causes of death in women of childbearing age from Las Tunas province were described; those caused by malignant tumors prevailed, being breast cancer the most common one (AU)

Factores de riesgo en el cáncer de pulmón / Risk Factors in the Lung Cancer

Introducción: el cáncer de pulmón es la neoplasia maligna más frecuente, y causante de un tercio de todas las muertes por cáncer, con un aumento significativo de su incidencia en los últimos años. Objetivo: identificar factores de riesgo asociados al cáncer de pulmón. Método: se realizó un estudio de casos y control en el Hospital Lucía Iñiguez Landín en el período entre julio de 2011 a enero de 2013. El universo estuvo constituido por 118 pacientes que ingresaron en Salas de Medicina que presentaban uno o varios factores de riesgo para esta enfermedad. De ellos, 59 pacientes con diagnósticos no relacionados con cáncer de pulmón fueron seleccionados como el grupo control y el grupo casos estuvo constituido por los 59 pacientes que ingresaron con este diagnóstico. Resultados: predominó en el grupo de casos el sexo masculino para 83,1%. Los fumadores representaron 71,2% del total del grupo casos. El hábito de fumar y la enfermedad pulmonar obstructiva crónica fueron los factores de riesgo de mayor importancia con una OR de 3,8 y 2,49, respectivamente. La exposición a sustancias cancerígenas y el alcoholismo no mostraron diferencias significativas entre ambos grupos. Conclusiones: el sexo masculino fue el más afectado por esta enfermedad. Los pacientes fumadores tuvieron 3,8 veces mayor riesgo de presentar cáncer de pulmón, que los no fumadores y los pacientes con enfermedad pulmonar obstructiva crónica presentaron 2,49 veces mayor probabilidad de desarrollar cáncer.

Introduction: the lung cancer is the most frequent malignant neoplasm and the cause of third of all deaths caused by the cancer, with a high increase of incidence in these last years. Objective: to determine the risk factors associated to lung cancer. Method: a cases and control study in Lucía Iñiguez Landín hospital from July 2011 to January 2013. The universe comprised 118 patients who were admitted in the medical - ward and presented one or some risks factors for this illness, 59 of these patients with diagnosis not related to lung cancer were selected as the control group and the case group was composed of 59 patients that were admitted with this diagnosis. Results: the male sex predominated in the case group the representing 83.1%. The smokers represented 71.2% of the cases group. The smoking and pulmonary obstructive chronic disease was the most important risk factors found with OR of 3.8 and 2.4 respectively. The cancerous substances exposure and the alcoholism did not show significant differences. Conclusion: the male sex was the most affected. The smokers had 3.8 more times of highest risk for lung cancer comparing with no smokers and those patients with obstructive pulmonary disease had 2.49 times of high probability for developing cancer.

Factores de riesgo en el cáncer de pulmón / Risk Factors in the Lung Cancer

Introducción: el cáncer de pulmón es la neoplasia maligna más frecuente, y causante de un tercio de todas las muertes por cáncer, con un aumento significativo de su incidencia en los últimos años.Objetivo: identificar factores de riesgo asociados al cáncer de pulmón.Método: se realizó un estudio de casos y control en el Hospital Lucía Iñiguez Landín en el período entre julio de 2011 a enero de 2013. El universo estuvo constituido por 118 pacientes que ingresaron en Salas de Medicina que presentaban uno o varios factores de riesgo para esta enfermedad. De ellos, 59 pacientes con diagnósticos no relacionados con cáncer de pulmón fueron seleccionados como el grupo control y el grupo casos estuvo constituido por los 59 pacientes que ingresaron con este diagnóstico.Resultados: predominó en el grupo de casos el sexo masculino para 83,1 por ciento. Los fumadores representaron 71,2 por ciento del total del grupo casos. El hábito de fumar y la enfermedad pulmonar obstructiva crónica fueron los factores de riesgo de mayor importancia con una OR de 3,8 y 2,49, respectivamente. La exposición a sustancias cancerígenas y el alcoholismo no mostraron diferencias significativas entre ambos grupos.Conclusiones: el sexo masculino fue el más afectado por esta enfermedad. Los pacientes fumadores tuvieron 3,8 veces mayor riesgo de presentar cáncer de pulmón, que los no fumadores y los pacientes con enfermedad pulmonar obstructiva crónica presentaron 2,49 veces mayor probabilidad de desarrollar cáncer(AU)

Introduction: the lung cancer is the most frequent malignant neoplasm and the cause of third of all deaths caused by the cancer, with a high increase of incidence in these last years.Objective: to determine the risk factors associated to lung cancer.Method: a cases and control study in Lucía Iñiguez Landín hospital from July 2011 to January 2013. The universe comprised 118 patients who were admitted in the medical ward and presented one or some risks factors for this illness, 59 of these patients with diagnosis not related to lung cancer were selected as the control group and the case group was composed of 59 patients that were admitted with this diagnosis.Results: the male sex predominated in the case group the representing 83.1 percent. The smokers represented 71.2 percent of the cases group. The smoking and pulmonary obstructive chronic disease was the most important risk factors found with OR of 3.8 and 2.4 respectively. The cancerous substances exposure and the alcoholism did not show significant differences.Conclusion: the male sex was the most affected. The smokers had 3.8 more times of highest risk for lung cancer comparing with no smokers and those patients with obstructive pulmonary disease had 2.49 times of high probability for developing cancer(AU)

Resource Classification for Medical Questions.

We present an approach for manually and automatically classifying the resource type of medical questions. Three types of resources are considered: patient-specific, general knowledge, and research. Using this approach, an automatic question answering system could select the best type of resource from which to consider answers. We first describe our methodology for manually annotating resource type on four different question corpora totaling over 5,000 questions. We then describe our approach for automatically identifying the appropriate type of resource. A supervised machine learning approach is used with lexical, syntactic, semantic, and topic-based feature types. This approach is able to achieve accuracies in the range of 80.9% to 92.8% across four datasets. Finally, we discuss the difficulties encountered in both manual and automatic classification of this challenging task.

Preparing a collection of radiology examinations for distribution and retrieval.

OBJECTIVE: Clinical documents made available for secondary use play an increasingly important role in discovery of clinical knowledge, development of research methods, and education. An important step in facilitating secondary use of clinical document collections is easy access to descriptions and samples that represent the content of the collections. This paper presents an approach to developing a collection of radiology examinations, including both the images and radiologist narrative reports, and making them publicly available in a searchable database. MATERIALS AND METHODS: The authors collected 3996 radiology reports from the Indiana Network for Patient Care and 8121 associated images from the hospitals' picture archiving systems. The images and reports were de-identified automatically and then the automatic de-identification was manually verified. The authors coded the key findings of the reports and empirically assessed the benefits of manual coding on retrieval. RESULTS: The automatic de-identification of the narrative was aggressive and achieved 100% precision at the cost of rendering a few findings uninterpretable. Automatic de-identification of images was not quite as perfect. Images for two of 3996 patients (0.05%) showed protected health information. Manual encoding of findings improved retrieval precision. CONCLUSION: Stringent de-identification methods can remove all identifiers from text radiology reports. DICOM de-identification of images does not remove all identifying information and needs special attention to images scanned from film. Adding manual coding to the radiologist narrative reports significantly improved relevancy of the retrieved clinical documents. The de-identified Indiana chest X-ray collection is available for searching and downloading from the National Library of Medicine (http://openi.nlm.nih.gov/).

The role of fine-grained annotations in supervised recognition of risk factors for heart disease from EHRs.

This paper describes a supervised machine learning approach for identifying heart disease risk factors in clinical text, and assessing the impact of annotation granularity and quality on the system's ability to recognize these risk factors. We utilize a series of support vector machine models in conjunction with manually built lexicons to classify triggers specific to each risk factor. The features used for classification were quite simple, utilizing only lexical information and ignoring higher-level linguistic information such as syntax and semantics. Instead, we incorporated high-quality data to train the models by annotating additional information on top of a standard corpus. Despite the relative simplicity of the system, it achieves the highest scores (micro- and macro-F1, and micro- and macro-recall) out of the 20 participants in the 2014 i2b2/UTHealth Shared Task. This system obtains a micro- (macro-) precision of 0.8951 (0.8965), recall of 0.9625 (0.9611), and F1-measure of 0.9276 (0.9277). Additionally, we perform a series of experiments to assess the value of the annotated data we created. These experiments show how manually-labeled negative annotations can improve information extraction performance, demonstrating the importance of high-quality, fine-grained natural language annotations.

Automatic Extraction and Post-coordination of Spatial Relations in Consumer Language.

To incorporate ontological concepts in natural language processing (NLP) it is often necessary to combine simple concepts into complex concepts (post-coordination). This is especially true in consumer language, where a more limited vocabulary forces consumers to utilize highly productive language that is almost impossible to pre-coordinate in an ontology. Our work focuses on recognizing an important case for post-coordination in natural language: spatial relations between disorders and anatomical structures. Consumers typically utilize such spatial relations when describing symptoms. We describe an annotated corpus of 2,000 sentences with 1,300 spatial relations, and a second corpus of 500 of these relations manually normalized to UMLS concepts. We use machine learning techniques to recognize these relations, obtaining good performance. Further, we experiment with methods to normalize the relations to an existing ontology. This two-step process is analogous to the combination of concept recognition and normalization, and achieves comparable results.

Automatic Classification of Structured Product Labels for Pregnancy Risk Drug Categories, a Machine Learning Approach.

With regular expressions and manual review, 18,342 FDA-approved drug product labels were processed to determine if the five standard pregnancy drug risk categories were mentioned in the label. After excluding 81 drugs with multiple-risk categories, 83% of the labels had a risk category within the text and 17% labels did not. We trained a Sequential Minimal Optimization algorithm on the labels containing pregnancy risk information segmented into standard document sections. For the evaluation of the classifier on the testing set, we used the Micromedex drug risk categories. The precautions section had the best performance for assigning drug risk categories, achieving Accuracy 0.79, Precision 0.66, Recall 0.64 and F1 measure 0.65. Missing pregnancy risk categories could be suggested using machine learning algorithms trained on the existing publicly available pregnancy risk information.

Analyzing U.S. prescription lists with RxNorm and the ATC/DDD Index.

OBJECTIVES: To evaluate the suitability of the ATC/DDD Index (Anatomical Therapeutic Chemical (ATC) Classification System/Defined Daily Dose) for analyzing prescription lists in the U.S. METHODS: We mapped RxNorm clinical drugs to ATC. We used this mapping to classify a large set of prescription drugs with ATC and compared the prescribed daily dose to the defined daily dose (DDD) in ATC. RESULTS: 64% of the 11,422 clinical drugs could be precisely mapped to ATC. 97% of the 87,001 RxNorm codes from the prescription dataset could be classified with ATC, and 97% of the prescribed daily doses could be assessed. CONCLUSIONS: Although the mapping of RxNorm ingredients to ATC appears to be largely incomplete, the most frequently prescribed drugs in the prescription dataset we analyzed were covered. This study demonstrates the feasibility of using ATC in conjunction with RxNorm for analyzing U.S. prescription datasets for drug classification and assessment of the prescribed daily doses.

Severe postpartum hemorrhage from uterine atony: a multicentric study.

OBJECTIVE: Postpartum hemorrhage (PPH) is an important cause of maternal mortality (MM) around the world. Seventy percent of the PPH corresponds to uterine atony. The objective of our study was to evaluate multicenter PPH cases during a 10-month period, and evaluate severe postpartum hemorrhage management. STUDY DESIGN: The study population is a cohort of vaginal delivery and cesarean section patients with severe postpartum hemorrhage secondary to uterine atony. The study was designed as a descriptive, prospective, longitudinal, and multicenter study, during 10 months in 13 teaching hospitals. RESULTS: Total live births during the study period were 124,019 with 218 patients (0.17%) with severe postpartum hemorrhage (SPHH). Total maternal deaths were 8, for mortality rate of 3.6% and a MM rate of 6.45/100,000 live births (LB). Maternal deaths were associated with inadequate transfusion therapy. CONCLUSIONS: In all patients with severe hemorrhage and subsequent hypovolemic shock, the most important therapy is intravascular volume resuscitation, to reduce the possibility of target organ damage and death. Similarly, the current proposals of transfusion therapy in severe or massive hemorrhage point to early transfusion of blood products and use of fresh frozen plasma, in addition to packed red blood cells, to prevent maternal deaths.