Active bone marrow segmentation based on computed tomography imaging in anal cancer patients: A machine-learning-based proof of concept

PURPOSE

Different methods are available to identify haematopoietically active bone marrow (ActBM). However, their use can be challenging for radiotherapy routine treatments, since they require specific equipment and dedicated time. A machine learning (ML) approach, based on radiomic features as inputs to three different classifiers, was applied to computed tomography (CT) images to identify haematopoietically active bone marrow in anal cancer patients.

METHODS

A total of 40 patients was assigned to the construction set (training set + test set). Fluorine-18-Fluorodeoxyglucose Positron Emission Tomography (18FDG-PET) images were used to detect the active part of the pelvic bone marrow (ActPBM) and stored as ground-truth for three subregions: iliac, lower pelvis and lumbosacral bone marrow (ActIBM, ActLPBM, ActLSBM). Three parameters were used for the correspondence analyses between 18FDG-PET and ML classifiers: DICE index, Precision and Recall.

RESULTS

For the 40-patient cohort, median values [min; max] of the Dice index were 0.69 [0.20; 0.84], 0.76 [0.25; 0.89], and 0.36 [0.15; 0.67] for ActIBM, ActLSBM, and ActLPBM, respectively. The Precision/Recall (P/R) ratio median value for the ActLPBM structure was 0.59 [0.20; 1.84] (over segmentation), while for the other two subregions the P/R ratio median has values of 1.249 [0.43; 4.15] for ActIBM and 1.093 [0.24; 1.91] for ActLSBM (under segmentation).

CONCLUSION

A satisfactory degree of overlap compared to 18FDG-PET was found for 2 out of the 3 subregions within pelvic bones. Further optimization and generalization of the process is required before clinical implementation.

End-to-end test for lung SBRT: An Italian multicentric pilot experience

PURPOSE

Set up a lung SBRT end-to-end (e2e) test and perform a multicentre validation.

MATERIAL AND METHODS

A group of medical physicists from four hospitals and the Italian Institute of Ionizing Radiation Metrology designed the present e2e test. One sub-group set up the test, while another tested its feasibility and ease of use. A satisfaction questionnaire was used to collect user feedback. Each participating centre (PC) received the ADAM breathing phantom, a microDiamond detector and radiochromic films. Following the e2e protocol, each PC performed its standard internal procedure for simulating, planning, and irradiating the phantom. Each PC uploaded its planning and treatment delivery data in a shared Google Drive. A single centre analyzed all the data.

RESULTS

The e2e test was successfully performed by all PCs. Participants' comments indicated that ADAM was well suited to the purpose and the protocol well described. All PCs performed the test in static and dynamic modes. The ratio between measured and planned point dose obtained by PC1, PC2, PC3, PC4 was: 0.99, 0.96, 1.01 and 1.01 (static track) and 0.99, 1.02, 1.01 and 0.94 (dynamic track). The gamma passing rates (3 % global, 3 mm) between planned and measured dose maps were 98.5 %, 94.0 %, 99.1 % and 94.0 % (static track) and 99.5 %, 96.5 %, 86.0 % and 94.5 % (dynamic track) for PC1, PC2, PC3 and PC4, respectively.

CONCLUSIONS

An e2e test for lung SBRT has been proposed and tested in a multicentre framework. The results and user feedback prove the validity of the proposed e2e test.