The Performance of Computer-Aided Detection Digital Chest X-ray Reading Technologies for Triage of Active Tuberculosis Among Persons With a History of Previous Tuberculosis.

BACKGROUND: Digital chest X-ray (dCXR) computer-aided detection (CAD) technology uses lung shape and texture analysis to determine the probability of tuberculosis (TB). However, many patients with previously treated TB have sequelae, which also distort lung shape and texture. We evaluated the diagnostic performance of 2 CAD systems for triage of active TB in patients with previously treated TB. METHODS: We conducted a retrospective analysis of data from a cross-sectional active TB case finding study. Participants ≥15 years, with ≥1 current TB symptom and complete data on history of previous TB, dCXR, and TB microbiological reference (Xpert MTB/RIF) were included. dCXRs were evaluated using CAD4TB (v.7.0) and qXR (v.3.0). We determined the diagnostic accuracy of both systems, overall and stratified by history of TB, using a single threshold for each system that achieved 90% sensitivity and maximized specificity in the overall population. RESULTS: Of 1884 participants, 452 (24.0%) had a history of previous TB. Prevalence of microbiologically confirmed TB among those with and without history of previous TB was 12.4% and 16.9%, respectively. Using CAD4TB, sensitivity and specificity were 89.3% (95% CI: 78.1-96.0%) and 24.0% (19.9-28.5%) and 90.5% (86.1-93.3%) and 60.3% (57.4-63.0%) among those with and without previous TB, respectively. Using qXR, sensitivity and specificity were 94.6% (95% CI: 85.1-98.9%) and 22.2% (18.2-26.6%) and 89.7% (85.1-93.2%) and 61.8% (58.9-64.5%) among those with and without previous TB, respectively. CONCLUSIONS: The performance of CAD systems as a TB triage tool is decreased among persons previously treated for TB.

Clinical and radiographic characteristics of presumptive tuberculosis patients previously treated for tuberculosis in Zambia.

BACKGROUND: Persistent respiratory symptoms and radiographic abnormalities are common among individuals previously treated for tuberculosis (TB) and may contribute to misdiagnosis and incorrect treatment when they seek care. We sought to determine if clinical and radiographic characteristics differed among previously treated, presumptive TB patients according to their current TB disease status. METHODS: Adults (>18 years of age) seeking care at a public health facility in Lusaka, Zambia were systematically evaluated for active TB using symptom screening and chest X-ray. All patients with presumptive TB submitted a sputum sample for microbiological TB testing. Patients who reported a prior history of TB treatment were included in the present analysis. 'Confirmed TB' was defined by the detection of TB using Xpert Ultra and/or liquid culture, while 'possible TB' was defined by the receipt of TB treatment without a positive Xpert Ultra or culture result. We evaluated the positive predictive value (PPV) of clinical symptoms and radiographic features for active TB alone and in combination. RESULTS: Of 740 presumptive TB patients, 144 (19%) had been previously treated for active TB. Of these, 19 (13%) patients had confirmed TB, 14 (10%) had possible TB, and 111 (77%) had no pulmonary TB. Overall, 119 (83%) patients had ≥1 current respiratory symptom-this did not differ according to current TB disease classification (95%, 93%, 79%; p = 0.23). Sixty-one patients (56%) had radiographic abnormalities suggestive of active TB and such findings were more common among patients with confirmed or possible TB compared to those without TB (93%, 71%, vs. 47%; p = 0.002). Most patients (n = 91, 83%) had at least one radiographic abnormality-no difference by current TB classification was observed (93%, 100%, 79%; p = 0.08). The PPV of any current respiratory symptom, active TB radiographic finding, or any radiographic abnormality for TB was 13% (95%CI: 7-21%), 21% (95%CI: 12-34) and 14% (95%CI: 9-23), respectively; combining clinical and radiographic characteristics did not significantly improve the PPV for active TB. CONCLUSIONS: Among presumptive TB patients previously treated for TB, respiratory symptoms and radiographic abnormalities were common and poorly differentiated those with current active TB from those without current active TB. Reliance on clinical and radiographic characteristics alone in this patient population may result in substantial overtreatment and therefore, microbiological investigations should be used to inform TB treatment decisions whenever possible.

The use a virtual interactive system to enhance gynecologic oncology multi-disciplinary care in Zambia.

BACKGROUND: Gynecologic malignancies represent a significant proportion of the increasing cancer burden in Zambia. The care and outcomes of cancer patients improves in a multidisciplinary team but insufficient coordination in weak health systems is a barrier to this model of care. The World Health Organization (WHO) identifies digital health interventions as tools to bridge the gap between challenges and health systems' goals in low resource environments. A low-cost innovative virtual interaction system (VIS) was integrated into the gynecologic oncology multidisciplinary tumor board to enhance patient navigation and management. METHODS: Paper-based forms used by each specialist group participating in the tumor board were collected and placed on a digital platform. Tumor board members were interviewed to assess their acceptability of the new digital modules and ensure correct nomenclature was being used for data entry. This was followed by an orientation and launch of VIS. RESULTS: For a 12-week period following the launch, 197 gynecologic oncology patients were registered at the Zambian national cancer center (Cancer Diseases Hospital), of which 130 were entered in the VIS, and of those less than half had management decisions recorded. The median time from first visit to simulation for radiotherapy was 52 days; from simulation to start of treatment 102.5 days. Eighteen (14 %) of the 130 patients entered into the VIS were recorded as lost to follow up. CONCLUSION: The introduction and implementation of a digital intervention for navigation and management of gynecologic patients in a low resource environment proved both feasible and acceptable. It provides an instant easily accessible platform for important information on time intervals and delays in the patient pathway. End user orientation and support is an integral part to its successful integration and consistent management of the data is required to maximize its impact on efficiency, effectiveness and patient care.