Development of a patient-specific two-compartment anthropomorphic breast phantom

Standardized breast models often fail to capture the complex, patient-specific anatomy required for precise imaging assessments. That uncertainty drove the need for more realistic physical representations of internal tissue distributions. Prior research has shown that existing phantoms lack the necessary anatomical fidelity for evaluating pendant breast geometries. No prior work had resolved the challenge of creating custom models that accurately reflect individual glandular and adipose tissue arrangements. This gap motivated the development of a new construction technique using patient-specific data. Researchers have previously relied on generic geometries that do not account for natural breast deformation. That limitation hinders the evaluation of advanced diagnostic tools in clinical settings. This paper addresses these constraints by utilizing high-resolution imaging to guide the fabrication of a bespoke physical model.

The aim of this paper is to develop a technique for constructing a two-compartment anthropomorphic breast phantom. This model specifically replicates the pendant anatomy of an individual patient. The researchers sought to address the lack of realistic physical models for evaluating dedicated breast imaging systems. They identified a need for a platform that accurately represents internal glandular and adipose tissue distributions. This motivation drove the creation of a process using patient-specific computed tomography data. The study focuses on translating digital image segmentation into a physical, modular structure. By using polyethylene and water, the authors intended to mimic the radiographic properties of breast tissues. This work provides a foundation for more precise dosimetry and imaging performance assessments in clinical research.

Review Approach involves utilizing three-dimensional images acquired from a prototype dedicated breast scanner. The team segmented these images into distinct adipose and glandular tissue regions. They divided the data into sections matching the thickness of the polyethylene stock. A computer-controlled water-jet machine cut the outer edges and internal glandular structures from the material. The researchers stacked these segments and encased them within a thermoplastic skin. They filled the remaining internal spaces with water to complete the model. This design allows for the insertion of microcalcifications to test imaging capabilities. The approach also incorporates point dose deposition measurements during scanning to evaluate radiation exposure.

Key Findings From the Literature indicate a mean coefficient of determination of 0.881 between the grayscale profiles of the patient and the phantom. The power law exponent describing anatomical noise was identical between the two. Microcalcifications were successfully visualized within the phantom during scanning procedures. Real-time air kerma rates fluctuated in response to the specific breast anatomy. Point dose deposition was found to be 7.1 percent greater than the mean glandular dose on average. Affine registration confirmed that the tissue distribution in the phantom closely matched the original patient images. The modular design allows for the evaluation of single segments or the entire breast volume. These results demonstrate the utility of the phantom for both imaging and dosimetry applications.

The authors successfully established a method to produce a patient-specific breast model using computed tomography data. This synthesis suggests that the resulting platform provides a reliable tool for assessing imaging performance. The findings indicate that the model accurately replicates the internal tissue distribution observed in the original patient. The researchers propose that the modular design facilitates both localized and volumetric investigations. This work implies that the phantom serves as a robust base for future dosimetry evaluations. The evidence confirms that the model allows for the visualization of small structures like microcalcifications. The study demonstrates that the physical construction closely matches the anatomical noise characteristics of the actual breast. These results support the use of such custom models in optimizing diagnostic procedures for individual patients.