Presentation is loading. Please wait.

Presentation is loading. Please wait.

ASSESSMENT OF INTERNAL BRUISE VOLUME OF SELECTED FRUITS USING MR IMAGING Ta-Te Lin, Yu-Che Cheng, Jen-Fang Yu Department of Bio-Industrial Mechatronics.

Similar presentations


Presentation on theme: "ASSESSMENT OF INTERNAL BRUISE VOLUME OF SELECTED FRUITS USING MR IMAGING Ta-Te Lin, Yu-Che Cheng, Jen-Fang Yu Department of Bio-Industrial Mechatronics."— Presentation transcript:

1 ASSESSMENT OF INTERNAL BRUISE VOLUME OF SELECTED FRUITS USING MR IMAGING Ta-Te Lin, Yu-Che Cheng, Jen-Fang Yu Department of Bio-Industrial Mechatronics Engineering, National Taiwan University, Taipei, Taiwan, ROC

2 INTRODUCTION MR Imaging MR Imaging Internal Bruise Evaluation Internal Bruise Evaluation Bruise Detection Bruise Detection 3D Reconstruction 3D Reconstruction

3 OBJECTIVES To develop a measurement method based on MR imaging to compute the internal bruise volume of selected fruits. To develop a measurement method based on MR imaging to compute the internal bruise volume of selected fruits. To develop image segmentation methods for bruise detection in MR images. To develop image segmentation methods for bruise detection in MR images. To compare bruise volume estimation using conventional methods and the MR imaging method. To compare bruise volume estimation using conventional methods and the MR imaging method.

4 MATERIALS & METHODS MR image acquisition MR image acquisition Impact bruise Impact bruise Image segmentation of bruise region Image segmentation of bruise region Volume estimation Volume estimation

5 MAJOR PARAMETERS IN MR IMAGING MR IMAGE ACQUISITION TR TR TE TE NEX NEX FOV FOV Resolution Resolution Slice thickness Slice thickness

6 BRUKER S330 MR SCANNER (3.0 T) MR IMAGE ACQUISITION

7 MR IMAGES OF SELECTED FRUITS MR IMAGE ACQUISITION Apple TR = 4000 ms TE = 45 ms Peach TR = 5000 ms TE = 60 ms Mango TR = 4500 ms TE = 60 ms Plum TR = 5000 ms TE = 60 ms

8 DROP TEST IMPACT BRUISE h1h1 h2h2 fruit m

9 BRUISE VOLUME ESTIMATION IMPACT BRUISE

10 FULL DEPTH METHOD IMPACT BRUISE

11 ELLIPSOID METHOD IMPACT BRUISE

12 IMAGE SEGMENTATION Manual threshold Manual threshold Automatic threshold Automatic threshold  Successive-iteration method  Kapur et al. method  Moment preserving method

13 IMAGE SEGMENTATION BRUISE REGION

14 IMAGE SEGMENTATION AUTOMATIC THRESHOLD METHODS Successive-iteration method Kapur et al. method Moment preserving method

15 RESULTS MR images of internal bruise MR images of internal bruise Comparisons of estimation methods Comparisons of estimation methods Image segmentation Image segmentation Bruise volume vs. impact energy Bruise volume vs. impact energy 3D reconstruction of bruise volume 3D reconstruction of bruise volume

16 MR IMAGES OF INTERNAL BRUISE APPLE

17 PLUM

18 COMPARISONS OF ESTIMATION METHODS Full Depth Method vs. MRI Method APPLES

19 COMPARISONS OF ESTIMATION METHODS Ellipsoid Method vs. MRI Method APPLES

20 COMPARISONS OF ESTIMATION METHODS Full Depth Method vs. MRI Method PLUMS

21 COMPARISONS OF ESTIMATION METHODS Ellipsoid Method vs. MRI Method PLUMS

22 SUMMARY TABLE COMPARISONS OF ESTIMATION METHODS MethodApple SlopeInterceptR2R2 Relative Error (%) RMSE (mm 3 ) Full Depth1.03-50.920.5825.8±19.12602 Ellipsoid1.06214.90.6225.8±22.42547 MethodPlum SlopeInterceptR2R2 Relative Error (%) RMSE (mm 3 ) Full Depth1.67248.00.8990.7±69.92556 Ellipsoid1.39-1256.50.7953.2±40.91280

23 IMAGE SEGMENTATION Successive-iteration Method

24 IMAGE SEGMENTATION Kapur et al. Method

25 IMAGE SEGMENTATION Moment Preserving Method

26 IMAGE SEGMENTATION SUMMARY TABLE MethodPlum SlopeInterceptR2R2 Relative error (%) RMSE (mm 3 ) Successive iteration 1.17381.780.86 40.8±42.5 1143 Kapur et al.1.71-983.550.91 31.1±25.2 1691 Moment preserving1.35328.090.8954.6±36.01581

27 BRUISE VOLUME VS IMPACT ENERGY APPLES Ellipsoid MethodMR Imaging Method

28 BRUISE VOLUME VS IMPACT ENERGY PLUMS Ellipsoid MethodMR Imaging Method

29 3D RECONSTRUCTION ApplePlum PeachMango

30 3D RECONSTRUCTION Full Depth Ellipsoid

31 CONCLUSIONS n The MR imaging approach provides an more accurate method to access internal bruise of selected fruits. The volume estimation errors for apples were 25.8 ± 15.6% and 22.3 ± 12.3% using full depth method and ellipsoid method, respectively. For plums, the estimation error were 90.7 ± 15.6% and 70.0 ± 12.3%, respectively. The volume estimation errors for apples were 25.8 ± 15.6% and 22.3 ± 12.3% using full depth method and ellipsoid method, respectively. For plums, the estimation error were 90.7 ± 15.6% and 70.0 ± 12.3%, respectively. n The three automatic threshold methods tended to over estimate bruise area. For MR images of plums, the best method were successive-iteration method which yielded an estimation error of 40.8±42.5%. n The 3D-reconstructed bruise volume helps in visualizing the extent and modes of impact bruise for selected fruits.

32 THANK YOU 謝 謝 謝 謝


Download ppt "ASSESSMENT OF INTERNAL BRUISE VOLUME OF SELECTED FRUITS USING MR IMAGING Ta-Te Lin, Yu-Che Cheng, Jen-Fang Yu Department of Bio-Industrial Mechatronics."

Similar presentations


Ads by Google