Authors: Cabral, Ricardo 1 ; Carvoeiras, Pedro 2 ; Fatana, João 2, ; Alves, Rita 1. 1 Centro Hospitalar Lisboa Norte - Hospital de Santa Maria; 2 Medical Consult, SA;
Establish a method to correlate image noise with individual demographic characteristics of each patient ; Optimize acquisition parameters without losing diagnostic quality.
Data standardization in CT guaranties a correct image exposure a fixed amount of radiation attenuation; CT users don t always know when/how to lower mas/kvp.
Quality of the exam changes, differing the level of image noise, with patients of different body types; There has been a tendency to increase mas (dose) in order to lower image noise.
Technical / Demographical Parameters Age; Antero- posterior (AP) and lateral (LAT) diameter; Weight; KVp; mas; Pitch; Slice thickness; Kernel; Height; CTDIvol; DLP; Scan Length; SD; Exam Head Lumbar Spine Chest Abdomen Pelvis Abdomen +Pelvis Sample Weight Slice CTDIvol Noise kvp mas (Kg) (mm) (mgy) (HU) male 38 min 11 120 200 2 29.6 1.8 female 61 median 69 120 350 5 49.5 3.15 total 99 max 114 120 400 5 65.6 8.2 male 27 min 28 120 228 2 15.2 6.8 female 47 median 70 120 350 2 25.8 13.55 total 74 max 115 140 435 2 33.8 31.5 male 15 min 28 120 60 1 3.9 3.7 female 11 median 71 120 150 3 10.0 10.4 total 26 max 105 120 250 3 16.7 16.6 male 9 min 50 120 163 2 11.1 8 female 9 median 70 120 210 3 13.6 14.35 total 18 max 100 120 300 3 19.4 22.6 male 10 min 65 120 145 3 10.8 10.9 female 5 median 70 120 210 3 13.5 11 total 15 max 90 120 370 3 26.0 13.1 male 9 min 48 120 80 2 5.2 8.1 female 12 median 71 120 190 3 12.5 14.4 total 21 max 100 120 220 3 15.5 24.5 Philips equipments 16 and 64 slice Brilliance / 4 slice Mx8000.
Tomografia Computorizada TC 1 (16cortes) Pa#ent ID Região anatómica Sexo Idade Altura (m) Peso (kg) ) (mgy) DLP (mgy*cm) Modulação de Dose kv mas Espessura de corte (mm) Pitch da aquisição (cm) Kernel Espessura (cm) Largura (cm) ROI (SD)
The standard deviation, was used as a quantity marker. q One image is chosen, and a ROI is drawn with 2 to 5 cm 2, in a homogeneous region ;
The images were digitally manipulated adding different levels of noise, simulating an mas lowering: Adaptation of the used mas (pseudo mas) to reduce the dose: Head Lumbar Spine Chest Abdomen Pelvis 350 400 150 210 210 315 360 135 189 189 262 300 112 157 157 210 240 90 126 126
Image quality was determined by two Radiologists; q The noise was evaluated without knowledge of the dose and technical parameters. Criteria to evaluate Image Quality Anatomic Criteria Physic Criteria Diagnostic Value General Quality Evaluating system Image Quality Score Unacceptable 1 Sub - Acceptable 2 Acceptable 3 Good Quality 4 High Quality 5
Image test with diferent noise levels added, using the Image J software: 350mAs 315 mas (- 10%) 262 (- 25%) mas 210 mas (- 40%) SD +4% SD +10% SD +27% SD
Percentile 75 of local DRL's: Local DRL s European DRL s Exam CTDIvol CTDIw DLP CTDIw DLP Head 52 52 838 60 1050 Lumbar Spine 27 22 843 35 800 Chest 11 11 394 35 280 Abdomen 15 15 536 35 780 Pelvis 20 20 574 35 570 Abdomen+Pelvis 14 13 662 35 1350
Quantitative evaluation: relation between image noise with radiation dose and demographic factors (Abdomen) y = 0.7725x - 8.889 R² = 0.64641 Noise 25 20 15 10 5 Noise 25 20 15 10 5 0 22 27 32 37 42 y = 0.4583x + 2.2114 R² = 0.40152 Effective Diameter (cm) Noise 25 20 15 10 5 y = 0.1326x + 4.6594 R² = 0.30527 0 20 25 30 35 40 IMC 0 40 50 60 70 80 90 100 110 Weight (Kg)
Noise Quantitative evaluation: relation between image noise with radiation dose and demographic factors (Chest) y = 0.4207x + 0.2681 18 16 14 12 10 8 6 4 2 0 Noise R² = 0.60998 20 25 30 35 40 y = 0.446x - 2.5687 R² = 0.45663 18 16 14 12 10 8 6 4 2 0 20 25 30 35 40 Effective Diameter (cm) IMC Ruído 18 16 14 12 10 8 6 4 2 0 y = 0.1144x + 3.0894 R² = 0.5134 20 30 40 50 60 70 80 90 100 110 Peso (Kg)
Quantitative evaluation: relation between image noise with radiation dose and demographic factors (Lumbar Spine) y = 0.8141x - 11.105 R² = 0.6398 Noise 35 30 25 20 15 10 5 0 20 25 30 35 40 45 y = 0.6702x - 3.8176 R² = 0.48093 Effective Diameter (cm) y = 0.2524x - 4.0622 R² = 0.54915 35 35 30 30 25 25 Noise 20 15 Noise 20 15 10 10 5 5 0 20 25 30 35 40 45 50 IMC 0 20 40 60 80 100 120 Weight (kg)
Quantitative evaluation: relation between image noise with radiation dose and demographic factors (Head) 9 y = 0.3888x - 3.4105 R² = 0.08771 8 7 6 Noise 5 4 3 2 1 0 14 14.5 15 15.5 16 16.5 17 17.5 18 18.5 19 Effective Diameter (cm)
Relation of image noise with mas from PMMA phantoms SD (HU) 35 30 25 20 15 10 5 0 50 150 250 350 450 mas product (mas) Chest Head Abdomen Lumbar Spine Linear (Chest) Linear (Head)
Quality perception in test images with added noise Image Quality (Noise) Score Abdomen CT images Score 6 5 4 3 2 1 0 100 150 200 250 300 350 Pseudo mas Image Quality (Noise) Score 6 5 4 3 2 1 Lumbar Spine CT images Score 0 150 200 250 300 350 400 450 Pseudo mas Image Quality (Noise) Score 6 5 4 3 2 1 0 Chest CT images Score 50 70 90 110 130 150 170 Pseudo mas Def [20-24] cm Def [30-34] cm Def [25-29] cm Def >35 cm Def [20-25] cm Def [31-36] cm Def [26-30] cm Def >37 cm BMI [15-20[ Kg/m^2 BMI [25-30[ Kg/m^2 BMI [20-25[ Kg/m^2 BMI >=30 Kg/m^2
Test images evaluation with the added noise and mas reduction as proposed Lumbar Spine: Def(cm) Routine mas Target Score Noise Increased Modified mas Dose Reduction 20-25 300 4 8% 270 10% 26-30 400 4 19% 300 25% 31-36 400 4 8% 360 10% >37 435 4 0% 435 0% Abdomen : Def(cm) Routine mas Target Score Noise Increased Modified mas Dose Reduction 20-24 210 4 32% 126 40% 25-29 210 4 15% 158 25% 30-34 210 4 11% 189 10% >35 300 4 15% 225 25% Chest: BMI(kg/m^2) Routine mas Target Score Noise Increased Modified mas Dose Reduction 15-19 100 4 9% 75 25% 20-24 150 4 9% 135 10% 25-29 160 4 9% 144 10% >30 160 4 0% 160 0%
Dose reduction by exam type mas reduction in standard Abdomen protocol mas reduction in standard Chest protocol 40% 30% 20% 10% 0% [20-25[ [25-30[ [30-35[ 35 25% 20% 15% 10% 5% 0% [20-25[ [25-30[ [30-35[ 35 mas reduction in standard Lumbar Spine protocol 30% 20% 10% 0% [15-19[ [20-24[ [25-29[ >30
Results show that the dose can be individually adapted while acquiring a CT exam with good quality wanted noise levels; Dose adjustment depend on size/weight and age of each patient; The image analysis helped the radiologist to determine the dose adjustment that would produce good images.
The dose must be individually adapted, with the need to standardize protocols and technical parameters adapted to the patient demographic characteristics (BMI and Effective Diameter); Radiologists have different acceptability of noise each department needs to adapt their noise levels, never exceeding the DRL s.
THANK YOU
REFERENCES
CT Image Quality, Wil Reddinger, M.Sc., R.T.(R)(CT); OutSource, Inc; Apr 19988. ICRP, International Commission on Radiological Protection. Recommendations of the International Commission on Radiological Protection. ICRP Publication 60; 1991. ICRP, International Commission on Radiological Protection. Managing Patient Dose in Multi- Detector Computed Tomography (MDCT). ICRP Publication 102. Annals of the ICRP 37,;Elsevier, 2007. IAEA,dose reduction in CT while Maintaining Diagnostic Confidence: A Feasibility/Demonstration Study, Vienna Sep 2009. European guidelines on quality criteria for computed tomography. Luxembourg: office for official Publications of the European Communities; 2000. X Ray CT, Nick Keat, ImPACT ST. George's Hospital; http://www.impactscan.org/slides/xrayct/sld001.htm; accessed 29-07- 2013.
ACKNOWLEDGMENT
Serviço de Imagiologia Hospital de Santa Maria; Medical Consult.