17
Embedded Processing Portfolio for Ultrasound

Embedded Processing Portfolio for Ultrasound

  • Upload
    others

  • View
    5

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Embedded Processing Portfolio for Ultrasound

Embedded Processing Portfolio for Ultrasound

Page 2: Embedded Processing Portfolio for Ultrasound

High performance, programmable platform

Processor performancespeeds image analysis –faster, clearer results

Power/size efficientprocessors enable portability – point of care imaging

Programmable DSPs enable software upgrades –extend product life

Complete line of Code compatible devices –provides scalable platform to address portable through high-end cart-based systems

Page 3: Embedded Processing Portfolio for Ultrasound

TI Confidential – NDA Restrictions

TI DSP SolutionsTI DSP Enables

TI’s Medical Imaging ProcessorsMission: to enable performance imaging every doctor demands

C66x™

C64x™

Integra™

Advances in medical processing

Latest algorithmsImproved image quality

More accurate diagnosis

Emerging Features

Innovative applicationssafer, non-invasive

Real-time diagnosis

Product differentiation

• Longer battery life

• Smaller form factor

Complete product line of high performance DSPs

Software programmable Flexible

Adaptable

Upgradeable

Real-time performance 1.25GHz single core

320GMACs/160GFlops multi-core

Medical driven roadmap

Portable medical applications

Low power processors and SOCs

• Highly integrated• Power efficient• C64x+ & C66x core DSP• OS, GUI, MMI, display• High connectivity

3

Sitara™

C6-Integra™

Davinci™

Page 4: Embedded Processing Portfolio for Ultrasound

Ultrasound system – Where DSP fits

DSPRF Demodulation

B-Mode

Color Flow

Spectral Doppler

Scan Control

SOC (DSP+MPU)Scan Conversion

Speckle Reduction

System Control

O/S

Display

Storage

Page 5: Embedded Processing Portfolio for Ultrasound

Multi-core DSP

Console Ultrasound Solution

5

• FPGA beam forms and routes data to DSP via SRIO

• C6657 (2-core) for scan control & back end image processing algorithms.

• PC performs system control, MMI, interface to PACs

• Lower power & system cost solution

C6657 (~3.5w)

SPI

DDR

3

64-bit1.33GHz

DDR3

control

data

66x (x2)1GHzL1P: 32 KBL1D: 32KBL2: 512KB

Multicore Navigator

DAC

McB

SP

SR

IOx

4Probe

MSMC

Shared L2: 4MB

HL 50

MMIPACs i/fMode SwitchCineloop Mgmt

.

System Ctrl

lanesBeam-former

1 to 4

PCIe

AFE

DetectionCompression Velocity Est

Power EstScan Conv

FFTPeak/Mean Est

B-Mode Color Doppler

Scan Conv

Wall FilterRF Demod

.

Decimation

Speckle Red

PC

Page 6: Embedded Processing Portfolio for Ultrasound

Multi-core DSP

Portable Ultrasound Solution

6

• FPGA beam forms and routes data to DSP via SRIO

• C6657 (2-core) implements mid-back end image processing algorithms

• AM3874 performs system control, MMI, interface to PACs

• C6657 + AM3874 = ~4w total

C6657 (~3.5w)

SPI

DDR3

64-bit1.33GHz

DDR3

control

data

DDR 2/3

DISPLAY

Storage

PACs

AM 3874 (~1w)ARM®

Cortex™-A8800 MHz L1P: 32KBL1D: 32KBL2 : 256KB

SATA

DSS

GEm

ac

PCIe

32-bit333/666 MHz

66x (x2)1GHzL1P: 32 KBL1D: 32KBL2: 512KB

DDR

LCD

Multicore Navigator

Graphics Engine

Image Engine

USB

Printer

DAC

McB

SP

GPMC

Flash

Switch FabricSR

IOx

4

16-bit 200 MHz

Probe

MSMC

Shared L2: 4MB

HL 50

MMIPACs i/fMode SwitchCineloop Mgmt

.

System Ctrl

lanesBeam-former

1 to 4

PCIe

AFE

DetectionCompression Velocity Est

Power EstScan Conv

FFTPeak/Mean Est

B-Mode Color Doppler

Scan Conv

Wall FilterRF Demod

.

Decimation

Linux

.

O/S

AndroidQT

Speckle Red

DM814x

Page 7: Embedded Processing Portfolio for Ultrasound

TI Confidential – NDA Restrictions

SOC Based

Ultra-Portable Ultrasound Solution

7

DAC

control

data

DDR 2/3

DISPLAY

Storage

PACs

DM8148

ARM®

Cortex™-A81GHz L1P: 32KBL1D: 32KBL2 : 256KB

SATA

DSS

GEm

ac

PCIe

32-bit333/666 MHz

674x+750MHz L1P: 32KBL1D: 32KBL2 : 256KB

DD

R

LCD

Graphics Engine

Image Engine

USB

Printer

McA

SP

GPMC

Flash

Switch Fabric

16-bit 200 MHz

Probe

O/S MMIPACs i/fMode SwitchCineloop Mgmt

Speckle RedScan Conv

Velocity EstPower Est

.

Scan Conv

FFTPeak/Mean Est.

B-Mode Color DopplerFront End

BeamformingDetectionCompounding

System Ctrl

Wall Filter

FPGA

Beam-former

RF Demod

AFE

• FPGA beam-former + RF demod and routes data to SOC via PCIe

• ARM Cortex A8 performs system control, MMI, interface to PACs

• DSP performs back end image processing

• Video & Graphics h/w accelerator

AM387x

Page 8: Embedded Processing Portfolio for Ultrasound

TI Confidential – NDA Restrictions

Multi-core DSP Based Performance Ultrasound Solution

8

• FPGA beam forms and routes data to DSP via SRIO

• C6678 standard mid processing + 3D/4D, Elastography, & Speckle Reduction

• AM3874 performs system control, MMI, interface to PACs

control

data

Flash

64-bit1.33GHz

DDR3

DDR 2/3

DISPLAY

Storage

PACs

ARM®

Cortex™-A8800 MHz L1P: 32KBL1D: 32KBL2 : 256KB

SATA

DSS

GEm

ac

PCIe

32-bit333/666 MHz

DD

R

LCD

Graphics Engine

Image Engine

USB

Printer

DAC

GPMC

Switch Fabric

16-bit 200 MHz

Probe

MMIPACs i/fMode SwitchCineloop Mgmt

.

System Ctrl

lanesBeam-former

1 to 4

AFE

DetectionCompression Velocity Est

Power EstScan Conv

FFTPeak/Mean Est

B-Mode Color Doppler

Scan Conv

Wall FilterRF Demod

.

Linux

.

O/S

AndroidQT

C66x

SPI

DD

R3

66x (x8)1GHzL1P: 32 KBL1D: 32KBL2: 512KB

Multicore Navigator

SR

IOx

4

MSMC

Shared L2: 4MB

HL 50 PCIe x2

Mc BSP

3D/4DElastographySpeckle Red

AM 3874 DM814x

Page 9: Embedded Processing Portfolio for Ultrasound

9TI Confidential – NDA Restrictions 9

TI Confidential – NDA Restrictions

C6655/ C6657 (Sample Now!)

New C66x Core

– 2 C66x Cores @ 1.0GHz nominal (1.4GHz max)

– C6657 (2 Core), C6655 (1 core)

– High Performance Fixed & Floating Point DSP Cores

Power Optimized Design

– Target 3.5W for 2 Core, 2.5W for 1 Core implementation (85c case @ 1Ghz)

Keystone Multi-Core Architecture

– Multicore Navigator, TeraNet, Hyperlink

Memory Architecture

– 1MB Local L2 per Core

– 1MB Multicore Shared Memory (MSM)

– Boot ROM,DDR3-1600MHz (32-bit)

– Address Translation & ECC

Interfaces

– 4x RapidIO rev 2.1 (1x4, 2x2, 1x2+2x1)

– 2 lanes PCIe Gen II

– 10/100/1000 Mbps Ethernet SGMII ports

– Universal Parallel Port (16-bit) Muxed with EMIF -16

– I2C, SPI, 2x McBSP (Mux), 32 GPIO, 2 x UART, 4x Timers64, Semaphore

Other

– 2x VCP2, 1x TCP3d

– Multicore debugging (embedded trace per core / chip)

– 0.8 mm pitch flip chip package

– 21x21 package

– Ext Temp Range: -55C to 105C

– 40nm High Performance Node

– Smart reflex

Multicore Navigator

Te

raN

et

C66X

DSP

L1 L2

SRIO

x4

PCIe

x2

EMIF

16

I2C

SPIUART

Peripherals & IO

SGMII

2x VCP2

TCP3d

Communications

CoProcessors

Power Management

Debug

Multicore Shared Memory Controller

(MSMC)

Shared Memory 1MB

DDR3-

32b

EDMA

SysMon

System Elements

Memory Subsystem

Hyp

erL

ink

C66X

DSP

L1 L2

McBSP

uPP

Page 10: Embedded Processing Portfolio for Ultrasound

Shannon (TMS320C6678) – Block Diagram

10

Multicore Navigator

Te

raN

et

C66x

DSP

L1 L2

C66x

DSP

L1 L2

C66x

DSP

L1 L2

C66x

DSP

L1 L2

C66x

DSP

L1 L2

C66x

DSP

L1 L2

C66x

DSP

L1 L2

C66x

DSP

L1 L2

8 x CorePac

SRIO

x4

PCIe

x2

EMIF

16

TSIP

2x

I2C

SPIUART

Peripherals & IO

GbE

Switch

SGMIISGMII

IP Interfaces

Crypto

Packet

Accelerator

Network

CoProcessors

Power Management

Debug

Multicore Shared Memory Controller

(MSMC)

Shared Memory 4MB

DDR3-

64b

EDMA

SysMon

System Elements

Memory Subsystem

• Multi-Core KeyStone SoC

• Fixed/Floating CorePac• 8 CorePac @ 1.25 GHz

• 4.0 MB Shared L2

• 320G MAC, 160GFLOP, 60G DFLOPS

• ~10W

• Navigator• Queue Manager, Packet DMA

• Multicore Shared Memory

Controller• Low latency, high bandwidth memory access

• 3-port GigE Switch (Layer 2)

• PCIe gen-2, 2-lanes

• SRIO gen-2, 4-lanes

• HyperLink• 50G Baud Expansion Port

• Transparent to Software

Hyper

Link

50

Page 11: Embedded Processing Portfolio for Ultrasound

Switched Central Resource (SCR)

Peripherals

Memory Interfaces

DM8148

USB

2.0

x2

GPIOGMAC

Switch

PCIe McASP

x6

SPDIF

McBSP

I2C/

SPI

x4

UART

x6DCAN

x2

DDR3

x2

SDIO

/SD

x3

Async

EMIF/

NAND

SATA2

Display

On-Screen Display

Resizer

SD DAC (x2)

HDMI PHY

HD Video I/O (x2)

Video I/O

HD Video

Coprocessor

(x1)

3D Graphics

Engine

ARM micro-

processor

Fixed/Floating

point DSP

ARM

Cortex

A8 TM

C674x

DSP

Core

DM8148 ProcessorCores ARM Cortex A8™ (MPU) up to 1 GHz C674x™ Floating Point DSP Core up to 750 MHz

Memory ARM: 32KB L1I-Cache, 32KB L1 D-Cache, 256K L2 DSP: 32KB L1I-Cache, 32KB L1 D-Cache, 256K L2 Two DDR-800 Controllers

Coprocessors/Subsystem HD VICP 2.0 Accelerator at 320 MHz

– Real-Time HD Encode /Decode 3D Graphics engine Display Subsystem

Peripherals Gigabit EMAC Switch USB 2.0 Ctlr/PHY x 2 PCIe 2.0

SATA 3.0Gbps DDR3 – 800 x2 SD/SDIO x3 McASP x6, McBSP SPI, GPIO, I2C, UART, DCAN

Power Total Power – Typical <4 W

Package 23x23, 0.8mm pitch, 684 ball BGA

• Via Channels enable low cost design rules -- 4 mil traces and 10/20 mil escape vias

Back to: DM roadmap | product positioning

Page 12: Embedded Processing Portfolio for Ultrasound

Using DSPs & SOCs in Ultrasound Systems

• TI’s C6678 DSP with new C66x floating point core provides high

performance signal processing capabilities at low power

– C6657 dual core DSP can perform processing for mid-range system

– Upgrade to 4 or 8 core C6674/C6678 for more advanced algorithms

• C66x DSPs are well suited for processing such as:

– B-Mode (Detection, Compression)

– Color (Wall Filter, Velocity & Power estimation)

– Doppler (FFT, Peak Mean estimation)

– Scan Conversion

– 3D/4D, Elasto-graphy, Speckle Reduction

• Combining an FPGA for beam formation/routing and Sitara ARM SOC

C66x DSPs can provide a flexible, low-power solution for digital

ultrasound systems.

– High Level Language “Eclipse” development environment

– On-Chip DMA & Multicore Navigator for data movement.

– McBSP ports address I/O needs for CW Doppler & audio

Page 13: Embedded Processing Portfolio for Ultrasound

Using DSPs & SOCs in Portable Ultrasound Systems

• TI’s low-power Davinci SoCs allow flexibility on the back-end SoC for

various display options, image filtering and target identification on a

single chip:

– C674x DSP (fixed-/floating-point DSPs)

– Cortex-A8 for peripheral and communications control

– 3D graphics engine for rich UIs

– Rich display sub-system for multiple HD displays

– HD video encode/decode accelerators (Davinci devices only)

• High system connectivity with peripherals such as:

– Gigabit ethernet

– PCIe

– USB

– SATA

– SPI/GPIOs/more…

Page 14: Embedded Processing Portfolio for Ultrasound

LINUX

Scalable BSP/SDKReleases

ARM® MPU

ARM® MPU + DSP

ARM® MPU + DSP+ Video Accelerator Engine

DAVINCI™

C6-INTEGRA™

SITARA™

FREE Development license to use our Linux, Android, or WinCE Board FREESupport Packages (BSP) / Software Development Kits (SDK)

* For use on our ARM, ARM+DSP, and ARM+DSP+Multimedia Processors* Each release seamlessly and scalable works across all products

Complete DSP & ARM MPU Software Solutions by TI

Periph Libraries and Stacks

Multimedia Codecs

Middleware/Frameworks

Common IDE/Tools

Example SW & Demos

“Instant Expert” SW Development Kits

Page 15: Embedded Processing Portfolio for Ultrasound

Medical Software Toolkit 2.0

Optimized implementations of commonly used C64x+/C66x DSP processing blocks

Source Code:

Ultrasound:

• B-mode (Envelop Detection & Compression)

• DAS Receive Beam-forming

• Doppler Processing

• RF Demodulation and Decimation

• Scan Conversion

Optical Coherence Tomography

• Cubic Spline Interpolation

• Optimized FFT

3D Rendering

• Affine Warp

Request download at: http://focus.ti.com/docs/toolsw/folders/print/s2meddus.html

Page 16: Embedded Processing Portfolio for Ultrasound

TI OMAP3530 Mistral EVM

Medical Ultrasound Demo (MIDAS) Rev. 2All B-Mode, Color Flow, and Scan Conversion Processing on OMAP3530!

Display 640x480 @20fps

B-mode Estimation

Color Flow

Tissue Flow

& Blending

Scan

Conversion

Scan

ConversionFlow

Estimation

Wall

Filter

Ensemble

Aggregation

CompressionEnvelope

Detection

C64x+ DSP 430MHz

ARM Cortex A-8600MHz

DDR

• Runs Linux O/S

• User Interface , Control, Display

• Runs Ultrasound Algorithms

Input Data Size (Post RF Demod) Scan

Lines

Samples/

Scan Line

Bytes/

Sample

Ensemble kB/

frame

B-mode + Scan Conversion 128 416 4 - 208

Color Flow + Scan Conversion 64 256 4 8 512

Loading DSP ARM ms/fm

B-Mode 19% 6% 15

B-Mode+

Color Flow46% 21% 28

https://gstreamer.ti.com/gf/project/med_ultrasound/

Page 17: Embedded Processing Portfolio for Ultrasound

TI Confidential – NDA Restrictions

Medical Imaging DSP Value Proposition

C64x+™

Sitara™

C66x™

New & innovative algorithms in softwareImproved image quality & emerging features

Field upgrades, Flexibility, Adaptive coding

Deterministic signal processing architectureSupports latest real-time O/S for predictable & reliable performance

Portable imaging applicationsLow power SOC’s replace PC. (DSP+ARM, Graphics, Video accl…)

Longer battery life. Smaller form factor.

Scalable platforms: PortableValuePremiumCode compatible family of products

R&D SavingsReuse (code, hardware, development environment)

No hardware spins, eco’s, & timing closure bottlenecks

Time to MarketState of the art development tools: (Compilers, trace, emulation)

Develop & debug in high level language

Imaglib & Medical Software toolkit, 3d parties

Roadmap & Product continuityLong term supplier, Full product line: Analog, DSP, Power, etc…

10GHz-320GMACs/160GFlop DSP’s today,

Application support (Field, Factory, domain white papers & app notes)

17

C6-Integra™

Davinci™