Speeding Up Medical Image Processing in MatLab
Author Information
Author(s): Michel Bister, CS Yap, KH Ng, CH Tok
Primary Institution: University of Malaya
Hypothesis
Good programming practices can significantly enhance the speed of medical image processing in MatLab.
Conclusion
Proper programming techniques can make MatLab competitive with C programming for speed in medical image processing.
Supporting Evidence
- Vectorization can enhance performance by a factor of 25:1 to 42:1.
- Pre-allocation can improve performance by a factor of 3:1 to 11:1.
- Specialization can lead to performance gains of 15:1 to 60:1 in bilinear interpolation.
Takeaway
This study shows that by using smart coding techniques, we can make MatLab run much faster for processing medical images, just like faster languages.
Methodology
The study illustrates the effects of vectorization, pre-allocation, and specialization on the speed of algorithms used in medical image processing.
Limitations
The study does not address all possible optimizations or the impact of memory limitations on processing speed.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website