MyMedR

Displaying all 4 publications

Abstract:

Sort:

Fulltext Evaluating online health information quality using machine learning and deep learning: A systematic literature review

Baqraf YKA, Keikhosrokiani P, Al-Rawashdeh M

Digit Health, 2023;9:20552076231212296.
PMID: 38025112 DOI: 10.1177/20552076231212296

BACKGROUND: Due to the large volume of online health information, while quality remains dubious, understanding the usage of artificial intelligence to evaluate health information and surpass human-level performance is crucial. However, the existing studies still need a comprehensive review highlighting the vital machine, and Deep learning techniques for the automatic health information evaluation process.
OBJECTIVE: Therefore, this study outlines the most recent developments and the current state of the art regarding evaluating the quality of online health information on web pages and specifies the direction of future research.
METHODS: In this article, a systematic literature is conducted according to the PRISMA statement in eight online databases PubMed, Science Direct, Scopus, ACM, Springer Link, Wiley Online Library, Emerald Insight, and Web of Science to identify all empirical studies that use machine and deep learning models for evaluating the online health information quality. Furthermore, the selected techniques are compared based on their characteristics, such as health quality criteria, quality measurement tools, algorithm type, and achieved performance.
RESULTS: The included papers evaluate health information on web pages using over 100 quality criteria. The results show no universal quality dimensions used by health professionals and machine or deep learning practitioners while evaluating health information quality. In addition, the metrics used to assess the model performance are not the same as those used to evaluate human performance.
CONCLUSIONS: This systemic review offers a novel perspective in approaching the health information quality in web pages that can be used by machine and deep learning practitioners to tackle the problem more effectively.
Revising and exploring the variations in methodologies for establishing the diagnostic reference levels for paediatric PET/CT imaging

Alhorani Q, Alkhybari E, Rawashdeh M, Sabarudin A, Latiff RA, Al-Ibraheem A, et al.

Nucl Med Commun, 2023 Nov 01;44(11):937-943.
PMID: 37615527 DOI: 10.1097/MNM.0000000000001748

PET-computed tomography (PET/CT) is a hybrid imaging technique that combines anatomical and functional information; to investigate primary cancers, stage tumours, and track treatment response in paediatric oncology patients. However, there is debate in the literature about whether PET/CT could increase the risk of cancer in children, as the machine is utilizing two types of radiation, and paediatric patients have faster cell division and longer life expectancy. Therefore, it is essential to minimize radiation exposure by justifying and optimizing PET/CT examinations and ensure an acceptable image quality. Establishing diagnostic reference levels (DRLs) is a crucial quantitative indicator and effective tool to optimize paediatric imaging procedures. This review aimed to distinguish and acknowledge variations among published DRLs for paediatric patients in PET/CT procedures. A search of relevant articles was conducted using databases, that is, Embase, Scopus, Web of Science, and Medline, using the keywords: PET-computed tomography, computed tomography, PET, radiopharmaceutical, DRL, and their synonyms. Only English and full-text articles were included, with no limitations on the publication year. After the screening, four articles were selected, and the review reveals different DRL approaches for paediatric patients undergoing PET/CT, with primary variations observed in patient selection criteria, reporting of radiation dose values, and PET/CT equipment. The study suggests that future DRL methods for paediatric patients should prioritize data collection in accordance with international guidelines to better understand PET/CT dose discrepancies while also striving to optimize radiation doses without compromising the quality of PET/CT images.
National and local diagnostic reference levels for adult18F-FDG and CT in Jordanian PET/CT: findings and implications in practice

Alhorani Q, Alkhybari E, Rawashdeh M, Sabarudin A, Latiff RA, Al-Ibraheem A, et al.

J Radiol Prot, 2024 Mar 07;44(1).
PMID: 38387102 DOI: 10.1088/1361-6498/ad2c62

This study aims to report the findings of Jordanian national diagnostic reference level (NDRL) survey for fluorodeoxyglucose (18F-FDG) and local diagnostic reference level (LDRL) of computed tomography (CT) used for attenuation correction and anatomical localisation (AC-AL); and AC and diagnostic CT (AC-DX) within the context of whole-body WB and half-body HB adult oncology PET/CT scanning. Two-structured questionnaires were prepared to gather the necessary information: dosimetry data, patient demographics, equipment specification, and acquisition protocols for identified18F-FDG PET/CT procedures. The NDRL and achievable dose were reported based on the 75th and 50th percentiles for18F-FDG administered activity (AA), respectively. The LDRL was reported based on the 50th percentile for (CTDIvol) and (DLP). Data from 562 patients from four Jordanian PET/CT centres were collected. The survey revealed that Jordanian NDRL for AA (303 MBq) was within the acceptable range compared to the published-peer NDRL data (240-590 MBq). However, the18F-FDG AA varied across the participated PET/CT centres. The reported LDRL CTDIvoland DLP of CT used for (AC-AL) was 4.3 mGy and 459.3 mGy.cm for HB CT scan range, and 4.1 mGy and 659.9 mGy.cm for WB CT scans. The reported LDRL for CTDIvoland DLP for HB CT was higher when compared with the United Kingdom (3.2 mGy and 310 mGy.cm). Concurrently, in the context of WB CT, the reported values (i.e. CTDIvol and DLP) were also higher than both Kuwait (3.6 mGy and 659 mGy.cm) and Slovenia (3.6 mGy and 676 mGy.cm). The reported HB CT(AC-DX) was higher than Nordic, New Zealand and Swiss NDRLs and for WB (AC-DX) CT it was higher than Swiss NDRLs. This study reported the first Jordanian NDRL for18F-FDG and LDRL for HB and WB CT associated with18F-FDG PET/CT scans. This data is useful for Jordanian PET/CT centres to compare their LDRL to the suggested DRLs and utilise it in the process of optimising CT radiation doses.
Investigating knowledge of DRLs, image quality and radiation dose in PET/CT and CT imaging among medical imaging professionals

Alhorani Q, Al-Ibraheem A, Rawashdeh M, Alkhybari E, Sabarudin A, A Latiff R, et al.

Heliyon, 2024 May 15;10(9):e30030.
PMID: 38707442 DOI: 10.1016/j.heliyon.2024.e30030

OBJECTIVE: To investigate the knowledge of diagnostic reference levels (DRLs), image quality, radiation dose and protocol parameters among Jordanian medical imaging professionals (MIPs) involved in PET/CT and CT scan procedures.
MATERIALS AND METHODS: A questionnaire was designed and distributed to MIPs in Jordan. The survey comprised four sections: demographic data, MIP knowledge on dose/protocol parameters, image quality, and DRLs. Statistical analyses were performed utilizing Pearson's correlation, t-tests, ANOVA, and linear regression, with a significance level of 95 % and a p-value threshold of <0.05.
RESULTS: The study involved 147 participants. Most respondents were male (76.2 %), and most were aged 26-35 years (44.2 %). Approximately 51 % held a bachelor's degree, and the most common range of experience was 3-5 years (28.6 %). Participants showed a moderate level of knowledge regarding dose and protocol parameters, with a mean score of 61.8 %. The mean scores for knowledge of image quality and DRLs were 45.2 % and 44.8 %, respectively. The age group of the MIPs and the total experience were found to have a significant impact on the knowledge of the dose and protocol parameters, as well as the DRLs. Additionally, experience was found to have a significant influence on knowledge of the dose and protocol parameters. The study revealed a positive and significant effect of MIPs' knowledge of dose/protocol parameters and image quality on their knowledge of DRLs.
CONCLUSIONS: This study indicates that professionals across five specialties who are engaged in PET/CT and CT imaging possess a moderate understanding of dosage and protocol parameters. However, there is a notable gap in knowledge regarding DRLs and image quality. To address this issue, it is recommended that MIPs actively engage in educational programs emphasizing exposure parameters and their impact on image quality. Additionally, access to comprehensive education and training programs will enable MIPs to grasp the complexities of DRLs and their implications, facilitating their implementation in clinical practice.