Browsing by Subject "X-ray"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    A novel phantom model for mouse tumor dose assessment under MV beams
    (Wolters Kluwer, 2011) Gossman, Michael S.; Das, Indra J.; Sharma, Subhash C.; Lopez, Jeffrey P.; Howard, Candace M.; Claudio, Pier P.; Radiation Oncology, School of Medicine
    In order to determine a mouse's dose accurately and prior to engaging in live mouse radiobiological research, a tissue-equivalent tumor-bearing phantom mouse was constructed and bored to accommodate detectors. Comparisons were made among four different types of radiation detectors, each inserted into the mouse phantom for radiation measurement under a 6 MV linear accelerator beam. Dose detection response from a diode, thermoluminescent dosimeters, and metal-oxide semiconductor field-effect transistors were used and compared to that of a reference pinpoint ionization chamber. A computerized treatment planning system was also directly compared to the chamber. Each detector system demonstrated results similar to the dose computed by the treatment planning system, although some differences were noted. The average disagreement from an accelerator calibrated output dose prescription in the range of 200-400 cGy was -0.4% ± 0.5 σ for the diode, -2.4% ± 2.6 σ for the TLD, -2.9% ± 5.0 σ for the MOSFET, and +1.3% ± 1.4 σ for the treatment planning system. This phantom mouse design is unique, simple, reproducible, and therefore recommended as a standard approach to dosimetry for radiobiological mouse studies by means of any of the detectors used in this study. The authors fully advocate for treatment planning modeling when possible prior to linac-based dose delivery.
  • Thumbnail Image
    Item
    MedShift: Automated Identification of Shift Data for Medical Image Dataset Curation
    (IEEE, 2023) Guo, Xiaoyuan; Wawira Gichoya, Judy; Trivedi, Hari; Purkayastha, Saptarshi; Banerjee, Imon; Biomedical Engineering and Informatics, Luddy School of Informatics, Computing, and Engineering
    Automated curation of noisy external data in the medical domain has long been demanding as AI technologies should be validated on various sources with clean annotated data. To curate a high-quality dataset, identifying variance between the internal and external sources is a fundamental step as the data distributions from different sources can vary significantly and subsequently affect the performance of the AI models. Primary challenges for detecting data shifts are – (1) access to private data across healthcare institutions for manual detection, and (2) the lack of automated approaches to learn efficient shift-data representation without training samples. To overcome the problems, we propose an automated pipeline called MedShift to detect the top-level shift samples and evaluating the significance of shift data without sharing data between the internal and external organizations. MedShift employs unsupervised anomaly detectors to learn the internal distribution and identify samples showing significant shiftness for external datasets, and compared their performance. To quantify the effects of detected shift data, we train a multi-class classifier that learns internal domain knowledge and evaluating the classification performance for each class in external domains after dropping the shift data. We also propose a data quality metric to quantify the dissimilarity between the internal and external datasets. We verify the efficacy of MedShift with musculoskeletal radiographs (MURA) and chest X-rays datasets from more than one external source. Experiments show our proposed shift data detection pipeline can be beneficial for medical centers to curate high-quality datasets more efficiently. The code can be found at https://github.com/XiaoyuanGuo/MedShift. An interface introduction video to visualize our results is available at https://youtu.be/V3BF0P1sxQE.