Browsing by Author "Zhang, Enming"

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Annotating and Detecting Topics in Social Media Forum and Modelling the Annotation to Derive Directions-A Case Study
    (Research Square, 2021) B., Athira; Jones, Josette; Idicula, Sumam Mary; Kulanthaivel, Anand; Zhang, Enming; BioHealth Informatics, School of Informatics and Computing
    The widespread influence of social media impacts every aspect of life, including the healthcare sector. Although medics and health professionals are the final decision makers, the advice and recommendations obtained from fellow patients are significant. In this context, the present paper explores the topics of discussion posted by breast cancer patients and survivors on online forums. The study examines an online forum, Breastcancer.org, maps the discussion entries to several topics, and proposes a machine learning model based on a classification algorithm to characterize the topics. To explore the topics of breast cancer patients and survivors, approximately 1000 posts are selected and manually labeled with annotations. In contrast, millions of posts are available to build the labels. A semi-supervised learning technique is used to build the labels for the unlabeled data; hence, the large data are classified using a deep learning algorithm. The deep learning algorithm BiLSTM with BERT word embedding technique provided a better f1-score of 79.5%. This method is able to classify the following topics: medication reviews, clinician knowledge, various treatment options, seeking and providing support, diagnostic procedures, financial issues and implications for everyday life. What matters the most for the patients is coping with everyday living as well as seeking and providing emotional and informational support. The approach and findings show the potential of studying social media to provide insight into patients' experiences with cancer like critical health problems.
  • Thumbnail Image
    Item
    A Novel Pipeline for Targeting Breast Cancer Patients on Twitter for Clinical Trial Recruitment
    (Office of the Vice Chancellor for Research, IUPUI, 2016-04-08) Sligh, Jon; Abedtash, Hamed; Yang, Mengye; Zhang, Enming; Jones, Josette
    Background and Preliminary Exploration: Breast cancer is the leading form of cancer in women, estimated to reach the incidence rate of 246,660 in 2016 in the US population. Scientist have developed new therapies for mitigating the disease and side effects in recent years through conducting randomized clinical trials as the gold standard clinical research method. However, recruiting individuals into clinical trials including breast cancer patients has remained a significant challenge. Our preliminary analysis on ClinicalTrial.gov registry showed that the majority of terminated clinical trials were due to recruitment challenges. Out of 525 terminated trials on breast cancer patients registered in the database, 230 (43.8%) of the terminations happened due to low or slow accrual, 34 (6.5%) due to lack of funding, and 31 (5.9%) due to toxicity concerns. Objectives: In this study, we developed and assess a scalable framework to identify Twitter users who have breast cancer based on personal health mentions on Twitter. In fact, we are looking for “fingerprints” of patients’ health status on Twitter, a microblogging social networking service. This method could provide a new avenue for contacting potential study candidates for recruitment. Methods: We analyzed the tweets of users who were following at least one of the top 40 twitter accounts where breast cancer patients gather. The rationale behind this approach is that cancer patients are following certain Twitter accounts to access support from other patients, doctors, or healthcare institutions. Consequently, these top twitter accounts provide a central point in which to find actual patients with breast cancer. We retrieved users’ tweets from Twitter API, and processed through the framework to match cancer relevant words and phrases individually and in combinations (caner, benign, malignant, etc.), possessive terms (I, my, has, have, etc.), and supporting attributes (mass, tumor, hair loss, etc.) to determine if the user has been diagnosed with cancer. The performance of the pipeline was measured in terms of sensitivity and specificity of detecting actual breast cancer patients. Results: We retrieved 25,870,106 tweets of 40 cancer community followers on Twitter. After excluding “retweets” and non-related breast cancer messages, we selected 81,429 tweets for further processing. The developed text processing pipeline could find total of 462 tweets based on the predefined sets of rules, representing 218 unique users. Our new method of Twitter data retrieval and text processing could identify breast cancer patients with remarkable sensitivity of 88.7% and specificity of 91.0%.
  • Thumbnail Image
    Item
    Towards the Creation of a Novel Career-Based Health Informatics (HI) Curriculum Assessment: Mapping HI Job Competencies to HI Curriculum Competencies
    Kulanthaivel, Anand; Zhang, Enming; Katta, Shilpa; Jones, Josette
    In order to best determine what competencies and skills are required for various careers in Health Informatics (HI) and create appropriately matched academic curricula accreditation recommendations, it is important to inventory current HI industry job requirements and posted curricula outcomes with respect to existing curriculum assessment frameworks. For this study, a Clinical Informatics-related career competency list as published by the American Nursing Association (ANA) [l] is used as a guiding framework to establish the competencies required in HI-related careers. The Commission on Accreditation for Health Informatics and Information Management Education (CAHIIM), in conjunction with the American Medical Informatics Association (AMIA), furthermore, have created a list of seven knowledge domains and skills relevant to HI curricula. [2] Anderson & Krathwohl [3] performed a review and revision of Bloom's classical taxonomy [4] of learning; it is thus possible to utilize the taxonomy's revision in tandem with the CAHIIM-AMIA knowledge domain to better merge concepts from academic education competencies with those required by real world HI-related careers.
  • Thumbnail Image
    Item
    What Do They Mean by "Health Informatics"? Health Informations Posts Compared to Program Standards
    (IOS Press, 2017) Jones, Josette F.; Zhang, Enming; Kulanthaivel, Anand; Katta, Shilpa; BioHealth Informatics, School of Informatics and Computing
    There is a lack of alignment between and within the competencies and skills required by health informatics (HI) related jobs and those present in academic curriculum frameworks. This study uses computational topic modeling for gap analysis of career needs vs. curriculum objectives. The seven AMIA-CAHIIM-accepted core knowledge domains were used to categorize a corpus of HI-related job postings (N = 475) from a major United States-based job posting website. Computational modeling-generated topics were created and then compared and matched to the seven core knowledge domains. The HI-defining core domain, representing the intersection of health, technology and social/behavioral sciences matched only 45.9% of job posting content. Therefore, the authors suggest that bidirectional communication between academia and industry is needed in order to better align educational objectives to the demands of the job market.