3D prostate histology image reconstruction: Quantifying the impact of tissue deformation and histology section location

摘要: Background: Guidelines for localizing prostate cancer on imaging are ideally informed by registered post‑prostatectomy histology. 3D histology reconstruction methods can support this reintroducing spatial information lost during processing. The need to register small, high‑grade foci drives a high accuracy. Accurate method design is impacted the answers following central questions of work. (1) How does tissue deform processing? (2) What misalignment sections induced microtome cutting? (3) choice model affect accuracy? Materials and Methods: Histology, paraffin block face magnetic resonance images were acquired 18 whole mid‑gland slices from six prostates. 7-15 homologous landmarks identified each image. Tissue deformation due processing was characterized using target registration error (TRE) after landmark‑based under four models (rigid, similarity, affine thin-plate-spline [TPS]). front faces quantified manually landmarks. impact TRE measured eight comprising one with without constraining slice faces. Results: Isotropic scaling improved mean 0.8‑1.0 mm (all results reported as 95% confidence intervals), while skew or TPS <0.1 mm. 1.1‑1.9° (angle) 0.9‑1.3 (depth). Using isotropic scaling, constraint raised 0.6‑0.8 Conclusions: For sub‑millimeter accuracy, should not constrain be flexible enough scaling.