Georgios Kalaitzakis1, Themistoklis Boursianis2, Georgios Gourzoulidis3, Sofia Gourtsoyianni4, Georgia Lymperopoulou4, Konstantinos Marias5, Apostolos Karantanas6, Thomas G Maris7. 1. Department of Medical Physics, University of Crete, Heraklion, Crete, Greece. Electronic address: giorgo.kalaitzakis@gmail.com. 2. Department of Medical Physics, University of Crete, Heraklion, Crete, Greece. 3. Research & Measurements Center of OHS Hazardous Agents, OHS Directorate, Hellenic Ministry of Labor, Athens, Greece; Lighting Laboratory, National Technical University of Athens, Greece. 4. School of Medicine, National and Kapodistrian University of Athens, Areteion Hospital, Athens, Greece. 5. Computational BioMedicine Laboratory, Institute of Computer Science, Foundation for Research and Technology-Hellas (FORTH), Heraklion, Crete, Greece; Technological Educational Institute of Crete, Department of Informatics Engineering, Heraklion, Greece. 6. Department of Radiology, University of Crete, Heraklion, Crete, Greece; Computational BioMedicine Laboratory, Institute of Computer Science, Foundation for Research and Technology-Hellas (FORTH), Heraklion, Crete, Greece. 7. Department of Medical Physics, University of Crete, Heraklion, Crete, Greece; Computational BioMedicine Laboratory, Institute of Computer Science, Foundation for Research and Technology-Hellas (FORTH), Heraklion, Crete, Greece.
Abstract
PURPOSE: The aim of this study is to introduce a novel DWI-MRI phantom and to compare Apparent Diffusion Coefficient (ADC) measurements, utilizing EPI-DWI and HASTE-DWI sequences and two different fitting algorithms. MATERIALS AND METHODS: 23 test tubes with different sucrose concentrations and polyacrylamide gels were used as a phantom for ADC measurements. The phantom was scanned on a clinical MRI system (1.5 T) over a two-month period utilizing an EPI-DWI and a HASTE-DWI sequence. ADC maps were calculated using a Weighted Linear (WL) and a Non Linear (NL) fitting algorithm. Measurements were performed with two sequences and two fitting algorithms. Geometric Distortions (GD), Ghosting Ratios (GR) and Signal to Structured Noise Ratios (SSNRs) were estimated using both sequences from the resultant ADC parametric maps. RESULTS: Polyacrylamide gels reveal lower coefficient of variation (CV%) as compared to sucrose solutions. ADC measurements performed with WL and NL algorithms reveal identical results with both sequences. WL and NL algorithms require approx. 3 s and 7 min respectively, for a single slice. EPI-DWI reveals a mean percent ADC value difference of (+4.5%) as compared to HASTE-DWI, regardless the type of fitting algorithm. CONCLUSION: Polyacrylamide gels can serve as a better means for simulating ADC values, compared with sucrose solutions used in this study. WL can be proposed as the method for ADC measurements in daily clinical practice. WL is significantly faster than NL fitting method and equally precise. SSNR measured directly on ADC maps is an excellent means for testing the precision of ADC measurements.
PURPOSE: The aim of this study is to introduce a novel DWI-MRI phantom and to compare Apparent Diffusion Coefficient (ADC) measurements, utilizing EPI-DWI and HASTE-DWI sequences and two different fitting algorithms. MATERIALS AND METHODS: 23 test tubes with different sucrose concentrations and polyacrylamide gels were used as a phantom for ADC measurements. The phantom was scanned on a clinical MRI system (1.5 T) over a two-month period utilizing an EPI-DWI and a HASTE-DWI sequence. ADC maps were calculated using a Weighted Linear (WL) and a Non Linear (NL) fitting algorithm. Measurements were performed with two sequences and two fitting algorithms. Geometric Distortions (GD), Ghosting Ratios (GR) and Signal to Structured Noise Ratios (SSNRs) were estimated using both sequences from the resultant ADC parametric maps. RESULTS:Polyacrylamide gels reveal lower coefficient of variation (CV%) as compared to sucrose solutions. ADC measurements performed with WL and NL algorithms reveal identical results with both sequences. WL and NL algorithms require approx. 3 s and 7 min respectively, for a single slice. EPI-DWI reveals a mean percent ADC value difference of (+4.5%) as compared to HASTE-DWI, regardless the type of fitting algorithm. CONCLUSION:Polyacrylamide gels can serve as a better means for simulating ADC values, compared with sucrose solutions used in this study. WL can be proposed as the method for ADC measurements in daily clinical practice. WL is significantly faster than NL fitting method and equally precise. SSNR measured directly on ADC maps is an excellent means for testing the precision of ADC measurements.