Purpose To determine the most accurate similarity metric when using an independent system to verify automatically generated contours. Methods A reference autocontouring system (primary system to create clinical contours) and a verification autocontouring system (secondary system to test the primary contours) were used to generate a pair of 6 female pelvic structures (UteroCervix [uterus + cervix], CTVn [nodal clinical target volume (CTV)], PAN [para‐aortic lymph nodes], bladder, rectum, and kidneys) on 49 CT scans from our institution and 38 from other institutions. Additionally, clinically acceptable and unacceptable contours were manually generated using the 49 internal CT scans. Eleven similarity metrics (volumetric Dice similarity coefficient (DSC), Hausdorff distance, 95% Hausdorff distance, mean surface distance, and surface DSC with tolerances from 1 to 10 mm) were calculated between the reference and the verification autocontours, and between the manually generated and the verification autocontours. A support vector machine (SVM) was used to determine the threshold that separates clinically acceptable and unacceptable contours for each structure. The 11 metrics were investigated individually and in certain combinations. Linear, radial basis function, sigmoid, and polynomial kernels were tested using the combinations of metrics as inputs for the SVM. Results The highest contouring error detection accuracies were 0.91 for the UteroCervix, 0.90 for the CTVn, 0.89 for the PAN, 0.92 for the bladder, 0.95 for the rectum, and 0.97 for the kidneys and were achieved using surface DSCs with a thickness of 1, 2, or 3 mm. The linear kernel was the most accurate and consistent when a combination of metrics was used as an input for the SVM. However, the best model accuracy from the combinations of metrics was not better than the best model accuracy from a surface DSC as an input. Conclusions We distinguished clinically acceptable contours from clinically unacceptable contours with an accuracy higher than 0.9 for the targets and critical structures in patients with cervical cancer; the most accurate similarity metric was surface DSC with a thickness of 1, 2, or 3 mm.

PURPOSE Radiotherapy (RT) treatment at public hospitals in Nigeria is often interrupted by prolonged periods of machine breakdown because of insufficient funds for maintenance and repair. These delays have prompted the uptake of public-private partnerships (PPPs) to acquire and maintain RT equipment. This study aimed to understand Nigeria's current RT capacity and the impact of PPPs on RT availability and cost. METHODS Eleven radiation oncologists, each representing one of the 11 RT centers in Nigeria (eight public and three private), were invited to complete a survey on the type, status, acquisition, and maintenance plan of existing RT equipment, cost incurred by patients for external-beam radiation (EBRT) and brachytherapy treatment, and number of patients treated per year on each machine. Type and status of equipment at nonresponding facilities were obtained through literature review and confirmed with the facility. RESULTS A total of eight (81%) respondents completed the survey, all representing public centers, three of which reported PPP use. They reported 11 megavoltage units in total (seven linear accelerators [LINACs] and four Cobalt-60s) and 10 brachytherapy afterloaders. Of those, 57% (4/7) of the LINACs, 100% (4/4) of the Cobalt-60s, and 63% (7/11) of the afterloaders were in clinical use. All commissioned equipment supported by PPPs (three LINACs and one afterloader) were in operation. The public EBRT equipment were nonfunctional 35% of the year (resulting in 60% fewer patients treated per year). The PPP EBRT and afterloaders did not experience any periods of breakdown, but PPP costs were 338% higher than public equipment. CONCLUSION This study characterizes the use of PPP as a more reliable method of RT delivery in Nigeria, albeit at higher costs.

No abstract

Background African countries are underrepresented in cancer research, partly because of a lack of structured curricula on clinical research during medical education. To address this need, the MD Anderson and Zambia Virtual Clinical Research Training Program (MOZART) was developed jointly by MD Anderson Cancer Center (MDA) and the Cancer Diseases Hospital in Zambia (CDH) for Zambian clinical oncology trainees. We explored participant perspectives to provide insight for implementation of similar efforts. Materials and Methods The MD Anderson and Zambia Virtual Clinical Research Training Program consisted of weekly virtual lectures and support of Zambian-led research protocols through longitudinal mentorship groups that included CDH faculty and MDA peer and faculty mentors. Participants were contacted via email to take part in semi-structured interviews, which were conducted via teleconference and audio-recorded, transcribed, and coded. Emergent themes were extracted and are presented with representative verbatim quotations. Results Thirteen of the 14 (93%) trainees were interviewed. Emergent themes included (1) participants having diverse educational backgrounds but limited exposure to clinical research, (2) importance of cancer research specific to a resource-constrained setting, (3) complementary roles of peer mentors and local and international faculty mentors, (4) positive impact on clinical research skills but importance of a longitudinal program and early exposure to clinical research, and (5) challenges with executing research protocols. Conclusion To our knowledge, this is the first qualitative study of African clinical oncology trainees participating in a virtual clinical research training program. The lessons learned from semi-structured interviews with participants in MOZART provided valuable insights that can inform the development of similar clinical research training efforts and scale-up.