TEMPLATE DESIGN © 2008 www.PosterPresentations.com Project Goal: Pneumothorax Disease Diagnosis at Remote places /Battle field Diagnosing Method and Recorded Signals Pole Map Signal Reconstruction efficiency Disease Possible Sources & Existing Diagnosis facilities Treatment Medical Percussion Technique Signal Classification & Reorganization Problems Solution: Signal Decomposition 1. Non parametric Method: Fourier Transform (existing solution) in 1975 by Murray & Neilson 2. Parametric method: Supper Resolution Technique A given signal into a sum of simpler signals • One step process for finding poles as solving of eigen value problem • Computationally more efficient and consider per sample based analysis • Analyze the data on snapshot- by-snapshot basis • Less sensitive to noise. Assumption: Signal can be decomposed into a set of exponentially damped oscillation signal A= Amplitude, d=damping, φ=Initial Phase, ω=Angular frequency Mathematical Model of the signals Matrix Pencil Method ) 1 ( ) ( ) 1 ( ) 2 ( ) ( ) 1 ( ) 1 ( ) ( ) 2 ( ) 1 ( ) ( ) 1 ( ) 1 ( ) 0 ( ] [ + × − − − − − − + − = L L N N y N y L N y L N y L y L y y y L y L y y y Y [ ] 1 Y [ ] 2 Y Damped Resonant signals Reconstruction error Mode Freq [Hz] Damp (1/sec) Amplitude Phase (Red) 1 54.48771 66.89012 0.34 -0.35056 2 110.2193 121.512 0.38 0.702722 3 164.1704 95.30471 0.32 0.515389 4 214.5606 109.451 0.38 0.044167 5 266.1824 93.89007 0.18 0.016278 6 338.8896 44.42675 0.04 -0.29767 7 410.9627 57.9877 0.02 -0.8075 8 568.4136 439.7309 0.54 0.785278 9 695.767 47.45113 0.02 0.107944 10 798.962 377.5481 0.18 -0.44683 11 1080.323 252.2069 0.02 -0.36089 Mode Freq [Hz] Damp Phase (degree) Amp 1 16.04875 43.21943 178.06 0.02 2 100.1096 116.4388 42.52 0.3 3 225.7679 101.2071 -127.31 1.02 4 291.3409 276.7678 60.88 0.94 5 501.5601 234.0484 -68.93 0.14 6 645.7305 64.4511 160.23 0.04 7 813.0717 120.9754 102.05 0.02 8 1059.815 61.93891 37.43 0.02 Results: 4-Parameter Values Tympanic Signal 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -1.5 -1 -0.5 0 0.5 1 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 time (secondes) 1 2 3 4 5 6 7 9 10 11 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 time (secondes) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 time (secondes) Damped Tympanic signals Resonant signal pole placement Tympanic signal pole placement Reconstruction error 0 0.05 0.1 0.15 0.2 0 5 10 15 RMS Error Number of poles 0 0.05 0.1 0.15 0.2 0.25 0.3 0 2 4 6 RMS error Number of poles Resonant signal Tympanic signal Resonant: 35 % of poles are able to reconstruct the original signal Tympanic: 60 % of poles are able to reconstruct the original signal Reconstructed Signals Conclusion & Future Research Resonant Signal Moinuddin Bhuiyan, Eugene Malyarenko and Roman Maev Methods Resonant Tympanic Signals are short duration, non deterministic highly damped, random in nature, and Lack of descriptions Matrix Pencil Method 8 Data are used to form the Matrix to solve Non-linear problem as a linear problem Y=ZR 1 , , 1 , 0 , ) ( ) ( 1 − = + ≅ ∑ = N k kT n z R kT y S M i k i i S Vandermonde matrix represents the pole parameter Matrix pencil Formed Hankel Matrix is formed from data Y SVD Y’ Noise Filtration using Minimum Description Length (MDL) to Select the M ( Singular Values) [] [ ] [ ][ ] H V U Y ' ' Σ ≅ •Tympanic signal Can be recognized by less than 3-dominant damped Sinusoidal Signal • Resonant Signal needs 5 to 6 dominant damped Sinusoidal signals. • Tympanic signal indication comes in case of Pneumothorax disease • Harware-software co-design for real time implementation on Digital Circuit board •Penetrating chest Injury • Motor vehicle accident. • A gunshot wound Diagnosis •Chest X-Ray & CT-scan Resonant signals Tympanic signals Signals Poles measurement Signals amplitude and phase measurement Least Square Solution : a=z\y Acknowledgements Office of the Naval Research for Financial support Not possible to use in Remote areas or battle field