| Objective1.To assess the detection difference of iterative model reconstruction(IMR)and filtered back projection(FBP)technique for small liver lesions on thin precontrast CT images,and explore whether the IMR technique can supply better background images for small liver lesions puncture than that of FBP.2.A new near infrared optical navigation system was used to real-time simulated CT-guided puncture of small targets in static agar model,and assess the feasibility and accuracy of the optical navigation system for puncture small targets of ideal puncture model.3.To assess the feasibility and accuracy of the optical navigation system for puncture small targets of simulated hold-breathing and free-breathing model,and attempt to solve several critical problems of puncture small targets in static and dynamic model.4.To assess the feasibility and accuracy of the optical navigation system for puncture small targets of simulated hold-breathing and free-breathing miniature pig,discover and attempt to solve these new problems for living animal puncture.Part ?Comparison of IMR and FBP in detecting small liver metastases on thin precontrast CT images.1 Materials and Methods89 patients with liver metastases were enrolled in this study,all of them were scanned with conventional CT protocol,and all raw data were reconstructed into 1mm FBP and IMR images.Total number(TN)of small lesion(?10mm)were defined as the number of small lesions found on 1mm enhanced IMR images;Detection number(DN)of small lesion were defined as the number of small lesions found on 1mm plain IMR images or FBP images;Detection rate(DR)=DN/TN.The subjective quality of 1mm plain FBP/IMR images were evaluated with a five-scale.Calculate the contrast-to-noise ratio(CNR)of portal vein and inferior vena cava.2.ResultsA total of 239 small liver lesions were be found on 1mm enhanced IMR images(TN=239),and the DR value were 43.93%(105/239),17.57%(42/239)on 1mm plain IMR and FBP images relatively.The subjective score of MR and FBP images were 4.17±0.51?3.28±0.54,relatively(p<0.05).The CNR value of IMR images for portal vein and inferior vena cava are higher than that of FBP images.3.ConclusionThe DR value and overall image quality of 1mm plain CT images with IMR was significantly better than with FBP technique,and the results of our study confirmed that the IMR technique can supply better thin slice background image for small liver lesions puncture than that of FBP technique.Part ?Evaluation of the feasibility and accuracy of a new near infrared optical navigation system for real-time simulated CT-guided puncture of small targets in agar phantom.1 Materials and Methods1.1 Experiment designsAll objects were divided into two groups:conventional CT navigation group(conventional group)and near infrared optical navigation(navigation group),there were 60 small targets in each group.1.2 EquipmentAll CT scans performed with a 256-slice MDCT system.The following acquisition parameters were used:tube voltage/tube current,120kv/300mAs;FOV,350mmx350mm;matrix,512×512;collimation,128 ×0.625 mm;slice thickness/gap,1.0mm/0.5mm.Our partner provided the navigation system.1.3 puncture method1.3.1 Conventional group(1)The solid agar phantom with targets scanned with a CT to find out the target.Then,find out the corresponding surface area through the CT laser-positioning tool.(2)Attached the surface positioning tool,another CT scan performed to detect the target and surface positioning tool.In addition,choose the optimal puncture entrance,angle and depth.After that insert the biopsy needle until the needle tip arrive at the planned depth.(4)Scanned the phantom with CT,to display the relationship of the needle tip and target.1.3.2 Navigation group(1)The solid agar phantom with 7 reflective spheres was placed on the scanning bed,scanned the phantom with CT(2)The locations of these reflective spheres on CT images matched with that of the optical tracking system.(3)The calibration of the puncture needle performed,and relative information were transfer to navigation system,melt with agar phantom's information.(4)Find the optimal puncture section and entrance by these MPR images displayed on the screen,insert the biopsy needle until the needle tip arrive the center of the target.(5)Another CT scan performed to display the needle tip and the target.1.4 Image post-processing,observation and data collectionMultiplanar reformatted(MPR)images(Fig 1)reformatted to make sure the central axis of needle and the center of target displayed on the same plane.Accuracy data were as followings:(1)Target positioning error(TPE),defined as the distance between the needle tip and the target center,and the TE value?10mm was considered success puncture;(2)Lateral error(LE),defined as the shortest distance between the target center and the needle extended line;(3)Angular error(AE),defined as the angle between the needle and the planned trajectory;(4)Depth error(DE),defined as the difference between the planned trajectory depth(from the needle entrance to the target center)and true depth(from the needle entrance to the needle tip).Clinical feasibility data was evaluated with first success rate(FSR),defined as the ratio of success puncture number divided by total puncture number for the first time.1.4 Statistical analysisThe difference of FSR between conventional group and navigated group were compared with the Chi-square test,the difference of other data between the two groups were compared with the student t test,and a p value<0.05 was considered statistically significant.2.Resultsthe TPE value of conventional group and navigation group are 12.19±4.19mm,3.18±1.70mm,relatively(p<0.05);The LE value of conventional group and navigation group are 6.80±2.86mm,2.28±1.49mm,relatively(p<0.05);The DE value of conventional group and navigation group are2.66± 1.07mm,1.37±0.99mm,relatively(p<0.05);The AE value of conventional group and navigation group are 4.91±1.50°,2.96±1.29°,relatively(p<0.05);The FSR value of conventional group and navigation group are 35%(21/60),100%(60/60),relatively(p<0.05).3.ConclusionThe results of the study demonstrated that these accuracy data(including TPE,LE,AE,and DE)and clinical feasibility data(FSR)of navigation group are better than that of conventional group,which confirmed that the navigation system have great potential on static rigid model/organ.Further research is needed to evaluate the accuracy and clinical feasibility of puncture for small target in static or dynamic no-rigid model/organ.Part ?Evaluation of the feasibility and accuracy of a new near infrared optical navigation system for real-time simulated CT-guided puncture of small targets in exvivo porcine liver phantom(hold-breathing and free-breathing)1 Materials and Methods1.1 Experiment designs Hold-breathing phantom:Conventional group and navigation group.Free-breathing phantom:Conventional group and navigation group.45 targets per group..1.2 EquipmentCT scanner and scanning protocol,The same as Part II Navigation system:The hardware of the navigation system for Part II and Part III were the same one.In this part,we used the fiducial registration error(FRE)curve to real-time monitor the seven custom fiducial markers' location differences of different time.The wave trough of the FRE curve implies the smallest one.1.3 Puncture method:The same as Part II.1.4 Image Observation:The same as Part II.1.5 Statistical analysis:The same as Part II.2.ResultsHold-breathing phantom:the TPE value of conventional group and navigation group are10.69±3.38mm,5.61±3.23mm,relatively(p<0.05);The LE value of conventional group and navigation group are 8.31±3.66 mm,3.60±2.22 mm,relatively(p<0.05);The DE value of conventional group and navigation group are3.34±1.28mm,4.13±1.94mm,relatively(p<0.05);The AE value of conventional group and navigation group are 5.33±2.48°,4.37±2.82°,relatively(p<0.05);The FSR value of conventional group and navigation group are33%(15/45),87%(39/45),relatively(p<0.05).Free-breathing phantom:the TPE value of conventional group and navigation group arel3.47±3.50mm,9.17±3.77mm,relatively(p<0.05);the LE value of conventional group and navigation group are 11.56±3.45mm,8.13±3.52mm,relatively(p<0.05),the DE value of conventional group and navigation group are7.87±4.05°,6.38±3.61°,relatively(p>0.05);the DE value of conventional group and navigation group are4.26±2.20mm,3.75±1.61mm,relatively(p>0.05),the FSR value of conventional group and navigation group are11%(5/45)?62%(39/45),relatively(p<0.05)3.ConclusionThe results of the study demonstrated that these accuracy data(including TE and LE value)and the clinical feasibility data(FSR)of navigation group are better than that of conventional group on both hold-breathing and free-breathing liver phantoms.Further technical improvements and studies in animals are required.Part IVEvaluation of the feasibility and accuracy of a new near infrared optical navigation system for real-time simulated CT-guided puncture of small targets in porcine model.1 Materials and Methods1.1 Experiment designs Hold-breathing animal:Conventional group and navigation group.Free-breathing animal:Conventional group and navigation group.45 targets per group.1.2 Imaging methods and near infrared optical navigation system The same as Part ?1.3 Puncture method The same as Part ?1.4 Image Observation The same as Part ?1.5 Statistical analysis The same as Part ?2.ResultsHold-breathing animal:the TPE value of conventional group and navigation group are13.86±3.45mm,5.22±3.02mm,relatively(p<0.05);The LE value of conventional group and navigation group are 9.39±3.20mm,4.01±2.43mm,relatively(p<0.05);The DE value of conventional group and navigation group are1.85±1.09mm,3.90±2.53mm,relatively(p<0.05);The AE value of conventional group and navigation group are 5.33±2.48°,4.37±2.82°,relatively(p>0.05);The FSR value of conventional group and navigation group are 16%(7/45),91%(41/45),relatively(p<0.05).Free-breathing animal:the TPE value of conventional group and navigation group are16.35±3.07mm,7.42±3.56mm,relatively(p<0.05);the LE value of conventional group and navigation group are10.49±4.62mm,5.56±3.36mm,relatively(p<0.05),the DE value of conventional group and navigation group are 5.33±2.48°?4.37±2.82°,relatively(p<0.05)the DE value of conventional group and navigation group are2.89±1.56mm?3.90±3.28mm,relatively(p>0.05),the FSR value of conventional group and navigation group are 19%(4/45)?73%(33/45),relatively(p<0.05)3.ConclusionThe results of the study demonstrated that these accuracy data(including TE,LE,)and clinical feasibility data(FSR)of navigation group are better than that of conventional group on both hold-breathing animal and free-breathing animal.However,even with the help of FRE curve and the navigation system,it is still difficult to perform the puncture of the targets of free-breathing animal,further technical improvements are required. |
PDF Full Text Request