Browsing by Author "Mun, Seong K."

Now showing 1 - 7 of 7
  • Thumbnail Image
    Artificial Intelligence for the Future Radiology Diagnostic Service
    Mun, Seong K.; Wong, Kenneth H.; Lo, Shih-Chung B.; Li, Yanni; Bayarsaikhan, Shijir (Frontiers Media, 2021-01-28)
    Radiology historically has been a leader of digital transformation in healthcare. The introduction of digital imaging systems, picture archiving and communication systems (PACS), and teleradiology transformed radiology services over the past 30 years. Radiology is again at the crossroad for the next generation of transformation, possibly evolving as a one-stop integrated diagnostic service. Artificial intelligence and machine learning promise to offer radiology new powerful new digital tools to facilitate the next transformation. The radiology community has been developing computer-aided diagnosis (CAD) tools based on machine learning (ML) over the past 20 years. Among various AI techniques, deep-learning convolutional neural networks (CNN) and its variants have been widely used in medical image pattern recognition. Since the 1990s, many CAD tools and products have been developed. However, clinical adoption has been slow due to a lack of substantial clinical advantages, difficulties integrating into existing workflow, and uncertain business models. This paper proposes three pathways for AI’s role in radiology beyond current CNN based capabilities 1) improve the performance of CAD, 2) improve the productivity of radiology service by AI-assisted workflow, and 3) develop radiomics that integrate the data from radiology, pathology, and genomics to facilitate the emergence of a new integrated diagnostic service.
  • Thumbnail Image
    Automated Error Labeling in Radiation Oncology via Statistical Natural Language Processing
    Ganguly, Indrila; Buhrman, Graham; Kline, Ed; Mun, Seong K.; Sengupta, Srijan (MDPI, 2023-03-23)
    A report published in 2000 from the Institute of Medicine revealed that medical errors were a leading cause of patient deaths, and urged the development of error detection and reporting systems. The field of radiation oncology is particularly vulnerable to these errors due to its highly complex process workflow, the large number of interactions among various systems, devices, and medical personnel, as well as the extensive preparation and treatment delivery steps. Natural language processing (NLP)-aided statistical algorithms have the potential to significantly improve the discovery and reporting of these medical errors by relieving human reporters of the burden of event type categorization and creating an automated, streamlined system for error incidents. In this paper, we demonstrate text-classification models developed with clinical data from a full service radiation oncology center (test center) that can predict the broad level and first level category of an error given a free-text description of the error. All but one of the resulting models had an excellent performance as quantified by several metrics. The results also suggest that more development and more extensive training data would further improve future results.
  • Thumbnail Image
    Blockchain-Enabled Next Generation Access Control
    Dong, Yibin; Mun, Seong K.; Wang, Yue (Springer, 2022-01-01)
    In the past two decades, longitudinal personal health record (LPHR) adoption rate has been low in the United States. Patients’ privacy and security concerns was the primary negative factor impacting LPHR adoption. Patients desire to control the privacy of their own LPHR in multiple information systems at various facilities. However, little is known how to model and construct a scalable and interoperable LPHR with patient-controlled privacy and confidentiality that preserves patients’ health information integrity and availability. Understanding this problem and proposing a practical solution are considered important to increase LPHR adoption rate and improve the efficiency as well as the quality of care. Even though having the state-of-the-art encryption methodologies being applied to patients’ data, without a set of secure access control policies being implemented, LPHR patient data privacy is not guaranteed due to insider threats. We proposed a definition of “secure LPHR” and argued LPHR is secure when the security and privacy requirements are fulfilled through adopting an access control security model. In searching for an access control model, we enhanced the National Institute of Standards and Technology (NIST) next generation access control (NGAC) model by replacing the centralized access control policy database with a permissioned blockchain peer-to-peer database, which not only eases the race condition in NGAC, but also provides patient-managed access control policy update capability. We proposed a novel blockchain-enabled next generation access control (BeNGAC) model to protect security and privacy of LPHR. We sketched BeNGAC and LPHR architectures and identified limitations of the design.
  • Thumbnail Image
    A DICOM dataset for evaluation of medical image de-identification
    Rutherford, Michael; Mun, Seong K.; Levine, Betty; Bennett, William; Smith, Kirk; Farmer, Phil; Jarosz, Quasar; Wagner, Ulrike; Freyman, John; Blake, Geri; Tarbox, Lawrence; Farahani, Keyvan; Prior, Fred (2021-07-16)
    We developed a DICOM dataset that can be used to evaluate the performance of de-identification algorithms. DICOM objects (a total of 1,693 CT, MRI, PET, and digital X-ray images) were selected from datasets published in the Cancer Imaging Archive (TCIA). Synthetic Protected Health Information (PHI) was generated and inserted into selected DICOM Attributes to mimic typical clinical imaging exams. The DICOM Standard and TCIA curation audit logs guided the insertion of synthetic PHI into standard and non-standard DICOM data elements. A TCIA curation team tested the utility of the evaluation dataset. With this publication, the evaluation dataset (containing synthetic PHI) and de-identified evaluation dataset (the result of TCIA curation) are released on TCIA in advance of a competition, sponsored by the National Cancer Institute (NCI), for algorithmic de-identification of medical image datasets. The competition will use a much larger evaluation dataset constructed in the same manner. This paper describes the creation of the evaluation datasets and guidelines for their use.
  • Thumbnail Image
    Lessons learned from implementing the patient-centered medical home
    Green, Ellen P.; Wendland, John; Carver, M. Colette; Hughes Rinker, Cortney; Mun, Seong K. (Hindawi, 2012)
    The Patient-Centered Medical Home (PCMH) is a primary care model that provides coordinated and comprehensive care to patients to improve health outcomes. This paper addresses practical issues that arise when transitioning a traditional primary care practice into a PCMH recognized by the National Committee for Quality Assurance (NCQA). Individual organizations' experiences with this transition were gathered at a PCMH workshop in Alexandria, Virginia in June 2010. An analysis of their experiences has been used along with a literature review to reveal common challenges that must be addressed in ways that are responsive to the practice and patients' needs. These are: NCQA guidance, promoting provider buy-in, leveraging electronic medical records, changing office culture, and realigning workspace in the practice to accommodate services needed to carry out the intent of PCMH. The NCQA provides a set of standards for implementing the PCMH model, but these standards lack many specifics that will be relied on in location situations. While many researchers and providers have made critiques, we see this vagueness as allowing for greater flexibility in how a practice implements PCMH.
  • Thumbnail Image
    Perspective Chapter: Blockchain-Enabled Trusted Longitudinal Health Record
    Dong, Yibin; Mun, Seong K.; Wang, Yue (2022-09)
    In the United States, longitudinal personal health record (LPHR) adoption rate has been low in the past two decades. Patients’ privacy and security concern is a major roadblock. Patients like to control the privacy and security of their own LPHR distributed across multiple information systems at various facilities. However, little is known how a scalable and interoperable LPHR can be constructed with patient-controlled security and privacy that both patients and providers trust. As an effort to increase LPHR adoption rate and improve the efficiency and quality of care, we propose a blockchain-enabled trusted LPHR (BET-LPHR) design in which security and privacy are protected while patients have full control of the access permissions. Two limitations associated with the proposed design are discussed. Options and practical resolutions are presented to stimulate future research.
  • Thumbnail Image
    Special Issue: "Machine Learning for Computer-Aided Diagnosis in Biomedical Imaging"
    Mun, Seong K.; Koh, Dow-Mu (MDPI, 2022-05-27)
    The radiology imaging community has been developing computer-aided diagnosis (CAD) tools since the early 1990s before the imagination of artificial intelligence (AI) fueled many unbound healthcare expectations and other industries [...]