Basis for a hands free blood flow measurement with automated vessel focus

Basic Information

• Paper ID: 2510.11060
• Title: Basis for a hands free blood flow measurement with automated vessel focus
• Authors: R. Fuchs, N. Sumrah, J. Schwerdt, M. Unger, G. Stachel, M. Schultz, K. Lenk, T. Neumuth
• Classification: physics.med-ph eess.SP
• Institutions: Innovation Center for Computer-Assisted Surgery, University of Leipzig; Department of Cardiology, University Hospital Leipzig; GAMPT Company; Merseburg University of Applied Sciences
• Paper Link: https://arxiv.org/abs/2510.11060

Abstract

Cardiopulmonary resuscitation (CPR) is a critical intervention to ensure oxygen supply during cardiac arrest. However, the precise efficacy of chest compressions cannot currently be quantified, resulting in poor CPR quality even when performed by medical professionals. Quantification of blood flow via ultrasound Doppler measurement can provide guidance to emergency responders. This paper proposes a method addressing time constraints, limited anatomical knowledge, and system configuration limitations in emergency situations. An automated vessel identification method comprising three distinct phases was developed, utilizing a novel sensor probe with asymmetric angle piezoelectric ceramics for ultrasound measurement. The probe and prototype ultrasound hardware were employed in a pulsed-wave (PW) Doppler laboratory setting. Initial measurements qualitatively assessed blood flow to identify effective measurement locations on phantoms, followed by iterative depth-correlated frequency measurements to calculate blood flow periodicity and signal power scores. Extended monitoring mode was performed using the optimal scoring configuration. Blood flow values were compared with industrial flow sensor data, demonstrating an average coefficient of determination of 0.97 and average root mean square error of 3.84 ml/s. The proposed hardware and software solution provides a foundation for future development that may enable fully automated vessel identification during CPR.

Research Background and Motivation

Problem Definition

1. Inability to Quantify CPR Efficacy: The precise effect of chest compressions cannot currently be quantified in real-time, resulting in poor CPR quality
2. Absence of Cerebral Blood Flow Monitoring: Cerebral blood flow is a determinant of long-term survival, yet direct, non-invasive, and user-friendly measurement methods are lacking during resuscitation
3. Limitations of Existing Technology: While commercial ultrasound equipment can perform Doppler blood flow measurement, it lacks robust hands-free connection options during CPR and requires specialized knowledge

Research Significance

• European Cardiac Arrest Registry data show annual incidence of out-of-hospital cardiac arrest reaching 170 per 100,000 population
• Resuscitation success rates in emergency services are approximately 50-60%, but average hospital discharge survival rates are only 8%
• Inadequate CPR is a major cause of poor prognosis following return of spontaneous circulation
• Carotid artery blood flow monitoring can provide real-time feedback to emergency responders, improving CPR efficacy

Limitations of Existing Methods

1. FloPatch System: Uses continuous-wave Doppler, unable to precisely identify particle motion within vessels
2. DopplerRescue System: Although employing pulsed-wave Doppler, requires manual adjustment, impractical in emergency situations
3. Commercial Equipment: Lacks robust mounting options during CPR, requires professional skills for device parameterization and signal interpretation

Core Contributions

1. Proposed a three-phase automated vessel identification method: Complete workflow comprising initial mode, iterative mode, and monitoring mode
2. Developed an innovative sensor probe: Featuring three piezoelectric ceramic arrays with asymmetric angles (25°, 0°, -12°)
3. Implemented automated parameter optimization: Achieved automation through iterative sampling volume optimization and score calculation
4. Verified high-precision blood flow measurement: Achieved average coefficient of determination of 0.97 compared with industrial flow sensors
5. Provided a hands-free measurement solution: Established technical foundation for real-time blood flow monitoring during CPR

Methodology Details

Task Definition

Input: Ultrasound signals from the carotid artery region Output: Real-time blood flow velocity and volumetric flow rate Constraints: Automated configuration must be completed within 5 seconds, applicable to emergency rescue scenarios

System Architecture

Hardware Design

• Probe Structure: Polylactic acid (PLA) shell with polyetheretherketone (PEEK) filling for ultrasound propagation
• Three Piezoelectric Ceramics: Angles of 25°, 0°, and -12° relative to the probe surface normal vector
• Connection System: Connected to the main computational unit via three ultrasound modules, enabling independent triggering and measurement

Software Algorithm

1. Initial Mode

• Ultrasound emission at 50μs intervals using the 25° sensor
• Sampling volume: start time 5μs, duration 30μs
• Objective: Rapid assessment of vessel coverage within large sampling volumes

2. Iterative Mode

• Small sampling window depth scanning: starting at 5μs, incrementing by 2μs per step until 43μs
• Three sensors alternately serve as transmitters
• Score calculation: MGM power score combining blood flow periodicity and signal power

3. Monitoring Mode

• Long-term monitoring using optimal configuration
• Simultaneous reception by three sensors, generating three velocity curves
• Cross-calculation to compensate for unknown Doppler angle

Technical Innovations

1. Asymmetric Multi-Angle Design: Three fixed-angle sensors provide broader measurement configuration range
2. Automated Depth Identification: Automatic vessel location identification through iterative small sampling volumes
3. Intelligent Scoring System: Comprehensive scoring mechanism combining frequency domain analysis and power spectral density
4. Angle Compensation Algorithm: Doppler angle approximation utilizing fixed sensor angle relationships

Mathematical Model

Doppler Frequency Formula:

Δfi = f0 * (v/c) * (cos αSE + cos αEi)

Angle Compensation Calculation:

cos αSE = √(C²/(1 + C²))

where C is the frequency relationship constant between sensors.

Experimental Setup

Experimental Platform

• Phantom Fabrication: Agarose phantom containing 3D-printed blood vessel with 8mm diameter
• Circulation System: Mechanical pump, silicone tubing, filled with Doppler fluid
• Reference Standard: VISION 1005 2F66 flow sensor (Badger Meter)
• Simulation Parameters: Pulsatile blood flow at 60 BPM

Measurement Locations

• Locations 1 and 3: Proximal and mid-placement without vessel coverage
• Location 2: Central placement with complete vessel coverage

Evaluation Metrics

• Coefficient of Determination (R²): Assesses linear fit quality
• Correlation Coefficient: Pearson correlation coefficient
• Root Mean Square Error (RMSE): Quantifies prediction error

Experimental Results

Primary Results

Initial Mode Results

Iterative Mode Results

• Optimal Configuration: Sensor 3 as transmitter, Sensor 1 as receiver
• Optimal Depth: 21μs (15.75mm), score 1.22×10⁹
• Depth Range: Effective signal appears in 19-27μs (14.25-20.25mm)

Monitoring Mode Results

Key Findings

1. Position Sensitivity: Only location 2 demonstrated effective vessel coverage and Doppler signal
2. Depth Accuracy: Automatically identified vessel depth consistent with B-mode imaging results
3. High-Precision Measurement: Optimal sensor combination (T1T2) achieved R² of 0.98 and RMSE of 2.31 ml/s
4. Real-Time Performance: System capable of tracking dynamic changes across different flow states

Related Work

Major Research Directions

1. Continuous-Wave Doppler Methods: Such as FloPatch system, broad coverage but limited precision
2. Pulsed-Wave Doppler Methods: Such as DopplerRescue system, high precision but requires manual adjustment
3. Flexible Ultrasound Patches: Emerging technology but facing similar automation challenges

Advantages of This Work

• Compared to continuous-wave methods: Provides precise depth selection and angle compensation
• Compared to existing pulsed-wave methods: Achieves fully automated parameter optimization
• Compared to commercial equipment: Hands-free solution specifically designed for CPR scenarios

Conclusions and Discussion

Main Conclusions

1. Successfully verified feasibility of the three-phase automated vessel identification method
2. Asymmetric multi-angle sensor design effectively enhanced measurement adaptability
3. Automated parameter optimization algorithm accurately identified optimal measurement configuration
4. System achieved high-precision blood flow measurement under laboratory conditions

Limitations

1. Measurement Time: Iterative mode exceeds recommended 5-second configuration time
2. Ideal Conditions: Experiments conducted under near-ideal laboratory conditions
3. Absence of Clinical Validation: Not yet verified in real CPR scenarios
4. Anatomical Variation: Requires establishment of larger database to accommodate patient variability

Future Directions

1. Algorithm Optimization: Simplify analysis workflow to reduce configuration time
2. Machine Learning Integration: Develop intelligent recognition algorithms utilizing databases
3. Miniaturization Design: Develop portable devices integrable into cervical collars
4. Clinical Trials: Verify system performance in real CPR scenarios

In-Depth Evaluation

Strengths

1. Strong Innovation: Asymmetric multi-angle design and three-phase identification method demonstrate clear novelty
2. High Practical Value: Directly addresses clinical needs for CPR quality monitoring
3. Complete Technical Solution: Comprehensive solution from hardware design to algorithm implementation
4. Sufficient Verification: Multiple evaluation metrics and detailed ablation analysis

Weaknesses

1. Experimental Limitations: Verification only on phantoms, lacking human trial data
2. Time Efficiency: Iterative mode is time-consuming, not meeting emergency scenario requirements
3. Missing Cost Analysis: No device cost and feasibility analysis provided
4. Lack of Standardization: Scoring system threshold settings lack standardized basis

Impact

1. Academic Value: Provides new technical pathway for CPR blood flow monitoring research
2. Clinical Potential: Likely to significantly improve CPR quality and patient outcomes
3. Commercialization Prospects: Patents filed, possessing foundation for industrialization
4. Reproducibility: Detailed method description facilitates reproduction by other researchers

Applicable Scenarios

1. Out-of-Hospital Emergency Care: Provides real-time CPR quality feedback to emergency medical services
2. Hospital Emergency Departments: Assists clinicians in optimizing resuscitation strategies
3. Training and Education: Serves as objective assessment tool for CPR training
4. Research Applications: Provides new measurement modality for hemodynamic research

References

This paper cites 24 relevant references covering key research in CPR guidelines, blood flow monitoring technology, and ultrasound Doppler methods, providing solid theoretical foundation for the research.

Overall Assessment: This is a high-quality medical physics research paper proposing an innovative CPR blood flow monitoring solution. The technical methodology is novel, experimental verification is comprehensive, and it possesses significant clinical application value. Despite certain limitations, it opens new directions for field development and merits further clinical validation and commercialization exploration.