Browsing by Author "Griffith-Richards, Stephanie"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Open Access
    A semi-automatic technique to quantify complex tuberculous lung lesions on 18F-fluorodeoxyglucose positron emission tomography/computerised tomography images
    (Springer Berlin Heidelberg, 2018-06-25) Malherbe, Stephanus T; Dupont, Patrick; Kant, Ilse; Ahlers, Petri; Kriel, Magdalena; Loxton, André G; Chen, Ray Y; Via, Laura E; Thienemann, Friedrich; Wilkinson, Robert J; Barry, Clifton E; Griffith-Richards, Stephanie; Ellman, Annare; Ronacher, Katharina; Winter, Jill; Walzl, Gerhard; Warwick, James M
    Background There is a growing interest in the use of 18F-FDG PET-CT to monitor tuberculosis (TB) treatment response. However, TB causes complex and widespread pathology, which is challenging to segment and quantify in a reproducible manner. To address this, we developed a technique to standardise uptake (Z-score), segment and quantify tuberculous lung lesions on PET and CT concurrently, in order to track changes over time. We used open source tools and created a MATLAB script. The technique was optimised on a training set of five pulmonary tuberculosis (PTB) cases after standard TB therapy and 15 control patients with lesion-free lungs. Results We compared the proposed method to a fixed threshold (SUV > 1) and manual segmentation by two readers and piloted the technique successfully on scans of five control patients and five PTB cases (four cured and one failed treatment case), at diagnosis and after 1 and 6 months of treatment. There was a better correlation between the Z-score-based segmentation and manual segmentation than SUV > 1 and manual segmentation in terms of overall spatial overlap (measured in Dice similarity coefficient) and specificity (1 minus false positive volume fraction). However, SUV > 1 segmentation appeared more sensitive. Both the Z-score and SUV > 1 showed very low variability when measuring change over time. In addition, total glycolytic activity, calculated using segmentation by Z-score and lesion-to-background ratio, correlated well with traditional total glycolytic activity calculations. The technique quantified various PET and CT parameters, including the total glycolytic activity index, metabolic lesion volume, lesion volumes at different CT densities and combined PET and CT parameters. The quantified metrics showed a marked decrease in the cured cases, with changes already apparent at month one, but remained largely unchanged in the failed treatment case. Conclusions Our technique is promising to segment and quantify the lung scans of pulmonary tuberculosis patients in a semi-automatic manner, appropriate for measuring treatment response. Further validation is required in larger cohorts.
  • Thumbnail Image
    Item
    Open Access
    Reliability and diagnostic performance of CT imaging criteria in the diagnosis of tuberculous meningitis
    (Public Library of Science, 2012) Botha, Hugo; Ackerman, Christelle; Candy, Sally; Carr, Jonathan A; Griffith-Richards, Stephanie; Bateman, Kathleen J
    Introduction Abnormalities on CT imaging may contribute to the diagnosis of tuberculous meningitis (TBM). Recently, an expert consensus case definition (CCD) and set of imaging criteria for diagnosing basal meningeal enhancement (BME) have been proposed. This study aimed to evaluate the sensitivity, specificity and reliability of these in a prospective cohort of adult meningitis patients. METHODS: Initial diagnoses were based on the CCD, classifying patients into: ‘Definite TBM’ (microbiological confirmation), ‘Probable TBM’ (diagnostic score ≥10), ‘Possible TBM’ (diagnostic score 6-9), ‘Not TBM’ (confirmation of an alternative diagnosis) or ‘Uncertain’ (diagnostic score of <6). CT images were evaluated independently on two occasions by four experienced reviewers. Intra-rater and inter-rater agreement were calculated using the kappa statistic. Sensitivities and specificities were calculated using both ‘Definite TBM’ and either ‘Definite TBM’ or ‘Probable TBM’ as gold standards. RESULTS: CT scan criteria for BME had good intra-rater agreement (κ range 0.35-0.78) and fair to moderate inter-rater agreement (κ range 0.20-0.52). Intra- and inter-rater agreement on the CCD components were good to fair (κ  =  ranges 0.47-0.81 and 0.21-0.63). Using ‘Definite TBM’ as a gold standard, the criteria for BME were very specific (61.5%-100%), but insensitive (5.9%-29.4%). Similarly, the imaging components of the CCD were highly specific (69.2-100%) but lacked sensitivity (0-56.7%). Similar values were found when using ‘Definite TBM’ or ‘Probable TBM’ as a gold standard. DISCUSSION: The fair to moderate inter-rater agreement and poor sensitivities of the criteria for BME suggest that little reliance should be placed in these features in isolation. While the presence of the CCD criteria of acute infarction or tuberculoma(s) appears useful as rule-in criteria, their absence is of little help in excluding TBM. The CCD and criteria for BME, as well as any new criteria, need to be standardized and validated in prospective cohort studies.