Page 118 - GTM-4-3
P. 118

Global Translational Medicine                                      Rapid diagnostic imaging on biopsy needle



            testing was conducted using both ex vivo porcine tissue and   additional biopsies were left uncompressed for 2  min
            a murine tumor model. Fresh CNBs were obtained from   before fixation in 10% neutral-buffered formalin for 72 h.
            ex vivo porcine tissue, and the murine tumor FVB/N-Tg   A calibrated scale was integrated into the CoreView
            (TgMMTV-neu) mouse strain was used as a representative   fixture for murine biopsies, enhancing precision in
            model for human mammary tumors obtained from the   determining biopsy compression levels. This scale was
            Cancer Vaccine Institute in Seattle, Washington, using   designed based on the average height of a 14-gauge
            14-gauge needles. A total of approximately 20 samples were   CNB  (1.2  mm)  and featured  black  notches  spaced  at
            analyzed for each tissue type across different compression   50  µm increments. In the murine model, biopsies were
            levels.                                            compressed to 70%, 60%, 50%, and 40% of their original

              For  porcine  tissue,  biopsy  thickness  was measured   thickness. Before compression, these biopsies were stained
            on the biopsy gun using a caliper before compression.   with Hoechst and Rhodamine B solutions for 30 s, followed
            To prevent tissue dehydration, biopsies were showered   by rinsing with PBS. CNBs were then loaded onto the
            with PBS solution before being compressed. A  screw-  CoreView  demonstrator  for  controlled  compression  and
            based glass-slide compression device was used (Figure 4),   imaging using the MUSE microscope. Following imaging,
            consisting of two 3D-printed round disks, each marked   biopsies were fixed in 10% neutral-buffered formalin for
            with 16 evenly spaced reference points corresponding to   72 h before submission for histological processing.
            a 0.03215 mm increment of compression. The disks were   Following fixation, all specimens were submitted to the
            attached  to  two  M3  hex socket  screws  with  a  0.5  mm   UW Histology and Imaging Core for routine H&E staining
            pitch,  ensuring  uniform  compression  across  the  porcine   and imaging. A blinded histopathological evaluation was
            biopsy specimen. Biopsies from fresh pig breast tissue   conducted by a breast pathology specialist, who assessed
            were compressed to 50%, 40%, and 30% of their original   image sets corresponding to the control and compressed
            thickness, and calculated using Equation I.        conditions. Each set was evaluated for diagnostic quality
                    Original     Original thickness          and the presence of compression-induced artifacts to
                    thickness  −           %  compression×        determine the effects of controlled compression on biopsy
            #Marks   =                                 (I)     integrity.
                             0.03125 mm
                                                               2.4. Quantification of nuclear edge sharpness using
              For instance, a 1.2  mm thick biopsy required    ImageJ
            approximately 23 marks of screw rotation to achieve
            40% compression.  Table 1 shows this method and the   To quantify nuclear edge sharpness, grayscale 10× images
            corresponding thicknesses. Two biopsies were collected   of porcine tissue sections imaged using MUSE and
            for each compression condition, and compressed biopsies   conventional H&E brightfield images were analyzed in
            remained under applied pressure for 2 min before fixation   ImageJ. The scale was set using known reference length of
            in 10% neutral-buffered formalin. As controls, two   full porcine biopsies (~1 cm in length) and used to calibrate
                                                               the image scale, spanning a distance of 10,000 µm across
                                                               38,702 pixels. Using the Plot Profile tool, intensity values
                                                               were measured across the diameter of five representative
                                                               nuclei per imaging modality. For each profile, the minimum
                                                               and maximum grayscale intensities were recorded, and the
                                                               20% and 80% intensity levels from baseline were calculated.
                                                               The pixel distance between these two points was used as a
                                                               quantitative measure of how sharply intensity changed at
                                                               the nuclear boundary. Average distances were computed
                                                               for  each  modality to  compare  edge  gradients  between
                                                               MUSE and H&E images.

                                                               Table 1. Biopsy thickness under compression
                                                               Percentage of  Original thickness Target thickness Marks turned
                                                               compression
                                                               50         1.22 mm, 1.19 mm 0.61 mm, 0.60 mm  19.5, 19.2
            Figure  4.  Compression testing device. This simple device enables   40  1.12 mm, 1.22 mm 0.45 mm, 0.49 mm  21.4, 23.4
            quantitative assessment of tissue deformation while maintaining
            histopathological integrity.                       30         1.17 mm, 1.30 mm 0.35 mm, 0.39 mm  26.2, 29.1


            Volume 4 Issue 3 (2025)                        110                          doi: 10.36922/GTM025170039
   113   114   115   116   117   118   119   120   121   122   123