Reducing the question burden of patient reported outcome measures using Bayesian networks

2022-11-01
Yuceturk, Hakan
Gulle, Halime
TUNCER ŞAKAR, CEREN
Joyner, Christopher
Marsh, William
ÜNAL, EDİBE
Morrissey, Dylan
Yet, Barbaros
Patient Reported Outcome Measures (PROMs) are questionnaires completed by patients about aspects of their health status. They are a vital part of learning health systems as they are the primary source of information about important outcomes that are best assessed by patients such as pain, disability, anxiety and depression. The volume of questions can easily become burdensome. Previous techniques reduced this burden by dynamically selecting questions from question item banks which are specifically built for different latent constructs being measured. These techniques analyzed the information function between each question in the item bank and the measured construct based on item response theory then used this information function to dynamically select questions by computerized adaptive testing. Here we extend those ideas by using Bayesian Networks (BNs) to enable Computerized Adaptive Testing (CAT) for efficient and accurate question selection on widely-used existing PROMs. BNs offer more comprehensive probabilistic models of the connections between different PROM questions, allowing the use of information theoretic techniques to select the most informative questions. We tested our methods using five clinical PROM datasets, demonstrating that answering a small subset of questions selected with CAT has similar predictions and error to answering all questions in the PROM BN. Our results show that answering 30% -75% questions selected with CAT had an average area under the receiver operating characteristic curve (AUC) of 0.92 (min: 0.8 - max: 0.98) for predicting the measured constructs. BNs outperformed alternative CAT approaches with a 5% (min: 0.01% - max: 9%) average increase in the accuracy of predicting the responses to unanswered question items.
JOURNAL OF BIOMEDICAL INFORMATICS

Suggestions

THE RELATIONSHIP BETWEEN NONSPECIFIC INFLAMMATION MARKERS AND COMPUTERIZED TOMOGRAPHY IMAGING IN NON-TRAUMATIC ACUTE ABDOMEN
Kozaci, Nalan; Ay, Mehmet Oguzhan; Cetinkunar, Suleyman; Soker, Gokhan; Başbuğ Erkan, Berna Burçak; Celikdemir, Alper; Acehan, Selen; Gulen, Muge; Satar, Salem (2014-01-01)
Aims: The purpose of the study was to establish whether the patients presenting with abdominal pain should undergo to contrast enhanced CT (Computed tomography). For this purpose, we have investigated the possible relationship between CRP (C-reactive protein),WBC (White blood cell), neutrophil levels and the presence of pathological findings determined on abdominal pelvic CT images.
Improving the Computer-Aided Estimation of Ulcerative Colitis Severity According to Mayo Endoscopic Score by Using Regression-Based Deep Learning
Polat, Gorkem; Kani, Haluk Tarik; Ergenc, Ilkay; Alahdab, Yesim Ozen; Temizel, Alptekin; Atug, Ozlen (2022-11-01)
Background Assessment of endoscopic activity in ulcerative colitis (UC) is important for treatment decisions and monitoring disease progress. However, substantial inter- and intraobserver variability in grading impairs the assessment. Our aim was to develop a computer-aided diagnosis system using deep learning to reduce subjectivity and improve the reliability of the assessment. Methods The cohort comprises 11 276 images from 564 patients who underwent colonoscopy for UC. We propose a regression-based deep ...
Predicting intraoperative pain in emergency endodontic patients: clinical study
Yucel, Olga; Ekici, Mugem Asli; İlk Dağ, Özlem; İLHAN, MUSTAFA NECMİ; KAYAOĞLU, GÜVEN (2018-01-01)
This prospective observational study sought to investigate the incidence of intraoperative pain (IOP) among emergency endodontic patients and to construct an IOP prediction model that includes preoperative pain level (PPL). All patients who underwent emergency endodontic treatment at Gazi University, Ankara, Turkey, during the spring term of 2016 were considered for inclusion in the study. Demographic and clinical variables and PPL were recorded. Local anesthesia was provided to all patients before beginnin...
Reliability of a Novel Visual Feedback System Developed for Increasing the Efficiency of Posterior Pelvic Tilt Exercise
Mengi, Gonen; Gencay-Can, Asli; Demirsoy, Nesrin; Tomsuk, Emrah; Koku, Ahmet Buğra; Beyazova, Mehmet (2014-09-01)
Objectives: This study aimed to develop and assess the reliability of a visual feedback system intended to improve the patient's perception and the performance of posterior pelvic tilt [PPT] exercises. The feedback setup designed and constructed by the authors included a bed covered with a mat with three embedded force sensors corresponding to the lumbosacral region of the subject lying supine; a data acquisition card to collect and convert data; and two monitors, one above the subject and one on the physic...
Understanding the perception towards using mHealth applications in practice: Physicians' perspective
Sezgin, Emre; Özkan Yıldırım, Sevgi; Yıldırım, İbrahim Soner (2018-03-01)
The objective of this study was to investigate physicians' perceptions to use mobile health applications in practice, and to identify influencing factors to use the technology. An mHealth technology acceptance model was proposed (M-TAM), and a cross-sectional survey was implemented using structured questionnaire to collect data. Online tools were used for inviting participants (physicians) and data collection from Turkey. The data was analyzed using Confirmatory Factor Analysis (CFA) and Structural Equation...
Citation Formats
H. Yuceturk et al., “Reducing the question burden of patient reported outcome measures using Bayesian networks,” JOURNAL OF BIOMEDICAL INFORMATICS, vol. 135, pp. 0–0, 2022, Accessed: 00, 2023. [Online]. Available: https://hdl.handle.net/11511/101696.