1. Sonic Millip3De: Massively Parallel 3D Stacked Accelerator for 3D Ultrasound
Richard Sampson* Ming Yang† Siyuan Wei†
Chaitali Chakrabarti† Thomas F. Wenisch*
*University of Michigan †Arizona State University

2. Portable Medical Imaging Devices
Medical imaging moving towards portability
MEDICS (X-Ray CT) [Dasika ‘10]
Handheld 2D Ultrasound [Fuller ‘09]
Not just a matter of convenience
Improved patient health [Gunnarsson ‘00, Weinreb ‘08]
Access in developing countries
Why ultrasound?
Low transmit power [Nelson ‘10]
No dangers or side-effects

3. Handheld 3D Ultrasound 3D has numerous benefits over 2D
Easier to interpret images
Greater volumetric accuracy
… as well as many challenges
12k transducers, 10M image points
10-20x beyond state of the art
High raw data bandwidth (6Tb/s)
Major bottleneck in state of the art
Tight handheld power budget (5W)

4. Why a Custom Accelerator?
Software algorithms load/store intensive
von Neumann designs inefficient
Large system would require over 700 DSPs
General purpose CPUs even less efficient
Architecture
Energy/Scanline
(1 fps)
Single Core
Time/Scanline
Intel Core i7-2670
25.08J
4.46s
ARM Cortex-A8
33.04J
132.18s
TI C6678 DSP
2.84J
2.27s

5. Contributions Iterative delay calculation algorithm
Reduces storage by over 400x
Enables streaming data flow
Sonic Millip3De design
Leverages 3D die stacking technology
Transform-select-reduce accelerator framework
Power and image analysis of Sonic Millip3De
Negligible change in image quality
Able to meet 5W power budget by 11nm node

6. Outline Introduction Ultrasound background Algorithm design
System design
Sonic Millip3De
Select Sub-Unit
Results and analysis
Conclusions

7. Ultrasound: Transmit and Receive
Image
Space 𝜏
Receive Raw Channel Data
Receive
Transducer
Focal
Points
Transmit
Transducer

8. Ultrasound: Transmit and Receive𝜏

9. Ultrasound: Transmit and Receive𝜏

10. Ultrasound: Transmit and Receive𝜏

11. Ultrasound: Transmit and Receive𝜏

12. Ultrasound: Transmit and Receive𝜏

13. Ultrasound: Transmit and Receive𝜏

14. Ultrasound: Transmit and Receive𝜏

15. Ultrasound: Transmit and Receive𝜏

16. Ultrasound: Transmit and Receive𝜏

17. Ultrasound: Transmit and Receive𝜏

18. Ultrasound: Transmit and Receive𝜏

19. Ultrasound: Transmit and Receive𝜏

20. Ultrasound: Transmit and Receive𝜏
Each transducer stores array of raw receive data

21. Ultrasound: Image Reconstruction
Image reconstructed from data based on round trip delay

22. Ultrasound: Image Reconstruction
Images from each transducer combined to produce full frame

23. Delay Index Calculation
Iterate through all image points for each transducer and calculate delay index
Often done with lookup tables (LUTs) instead
50 GB LUT required for target 3D system 𝑃
𝜏 𝑃

24. Challenges of Handheld 3D Ultrasound
Delay index LUT requires too much storage
New iterative algorithm reduces necessary constant storage by 400x
Peak raw data bandwidth (6Tb/s) infeasible
Sub-aperture multiplexing reduces peak data rate, but requires more transmits
Handheld power budget very tight (5W)
3D stacked, highly parallel data streaming design reconstructs images efficiently

25. Iterative Delay Index Calculation
Deltas between adjacent focal points on a scanline form smooth curve
Fit piecewise quadratic approx. to delta function
Two sections sufficient for negligible error
One section is off by 40
Section 1
Section 2

26. Sub-aperture Multiplexing
Peak raw data bandwidth (6Tb/s) infeasible
Solution: sub-aperture multiplexing
Transmit multiple times from same location
Receive with subset of transducers (sub-aperture)
Sum images together
Prior work: reduce data rate
Our design: also reduces HW and power requirements

27. System Design

28. Sonic Millp3De comprises 1,024 parallel pipelines
System Design
Sonic Millp3De comprises 1,024 parallel pipelines

29. System Design: Transducers
Be sure to mention high power requirement
Interchangeable CMOS transducer layer; can use older process

30. System Design: ADC/Storage
6kB SRAMs
Separate storage layer to reduce wire lengths

31. System Design: Transform-Select-Reduce
Accelerator units in fast, low power process

32. Select Sub-Unit Design
Selects sample closest to each focal point using our algorithm

33. Select Sub-Unit Design
Section 1
Section 2
All delays for a scanline estimated using 9 constants

34. Select Sub-Unit Design
Section 1
Section 2
A(n+1)2 + B(n+1) + C = (An2 + Bn + C) + 2An + (A+B)
Adders calculate next iteration of quadratic approximation

35. Select Sub-Unit Design
Section 1
Section 2
Decrementor selects sample for next image focal point

36. Select Sub-Unit Design
Section 1
Section 2
Section decrementor indicates when to change constants

37. Outline Introduction Ultrasound background Algorithm design
System design
Sonic Millip3De
Select Sub-Unit
Results and analysis
Conclusions

38. System Parameters Parameters Value Sub-apertures 12 Transmit Sources16
Transmits per Frame
192
Transducers per Sub-aperture
1,024
Total Transducers
12,288
Storage per Transducer
4,096 x 12 bits
Focal Points per Scanline
4,096
Image Depth
6 cm
Image Angular Width
π/4
Sampling Frequency
40 MHz
Interpolation Factor 4x
Interpolated Sampling Frequency (fs)
160 MHz
Speed of Sound (tissue)
1,540 m/s
Target Frame Rate
1 fps
Standard aperture and parameters for 3D abdominal scans

39. Image Quality Comparison
Simulations using Field II [Jensen ‘92, ‘95]
Ideal
Our Design (12 bit)
11 bit
Explain CNR (standard metric, gives ratio of contrast between cyst and tissue and the noise in those regions)
Bits
Ideal 14 13 12 11 10
CNR
2.972
2.942
2.960
2.536
2.233
Our design has negligible difference from ideal system

40. Power Analysis and Scaling
Can meet 5W by 11nm node

41. Conclusions
3D die stacked Sonic Millip3De design is able to meet 5W power budget by 11nm
Algorithm/HW co-design enables order-of-magnitude gains
Power and output quality goals often in conflict
Need guidance from domain experts to balance
Architects have much to offer for application-specific system designs

42. Questions? Special thanks to: Brian Fowlkes Oliver Kripfgans
Ron Dreslinski