Comparison of Supervised and Unsupervised Approaches for the Generation of Synthetic CT from Cone-Beam CT.
Ontology highlight
ABSTRACT:
SUBMITTER: Rossi M
PROVIDER: S-EPMC8395013 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Background. The present study was undertaken to compare the Hounsfield Unit (HU) in computed tomography (CT) with the gray level in CBCT in human tissues. Methods. In this study, 25 different soft and hard tissues were evaluated in 21 patients. CBCT images were taken with Newtom VGi machine (Verona, Italy) and CT images were prepared with Somatom Sensation unit (Siemens, Germany). The HU values of soft and hard tissues were compared with the gray level values of CBCT images. Results. There was a strong correlation between the HU in CT and the gray level in CBCT in soft tissues (P<0.001, R2 = 0.85) and hard tissues (P<0.001, R2 = 0.74) and in general (P<0.001, R2 = 0.91). Conclusion. A high degree of agreement was seen between HU in CT and gray level in CBCT in both hard and soft tissues. Since the gray level in CBCT was similar to HU in CT and can be used as a parameter determine bone density in implant treatment and also to determine the bone type, the CBCT technique is recommended in such cases due to its low radiation dose, short time and low cost compared to CT.
| S-EPMC6904917 | biostudies-literature
Project description:PurposeTo provide initial assessment of image quality and dose for a cone-beam computed tomographic (CT) scanner dedicated to extremity imaging.Materials and methodsA prototype cone-beam CT scanner has been developed for imaging the extremities, including the weight-bearing lower extremities. Initial technical assessment included evaluation of radiation dose measured as a function of kilovolt peak and tube output (in milliampere seconds), contrast resolution assessed in terms of the signal difference-to-noise ratio (SDNR), spatial resolution semiquantitatively assessed by using a line-pair module from a phantom, and qualitative evaluation of cadaver images for potential diagnostic value and image artifacts by an expert CT observer (musculoskeletal radiologist).ResultsThe dose for a nominal scan protocol (80 kVp, 108 mAs) was 9 mGy (absolute dose measured at the center of a CT dose index phantom). SDNR was maximized with the 80-kVp scan technique, and contrast resolution was sufficient for visualization of muscle, fat, ligaments and/or tendons, cartilage joint space, and bone. Spatial resolution in the axial plane exceeded 15 line pairs per centimeter. Streaks associated with x-ray scatter (in thicker regions of the patient--eg, the knee), beam hardening (about cortical bone--eg, the femoral shaft), and cone-beam artifacts (at joint space surfaces oriented along the scanning plane--eg, the interphalangeal joints) presented a slight impediment to visualization. Cadaver images (elbow, hand, knee, and foot) demonstrated excellent visibility of bone detail and good soft-tissue visibility suitable to a broad spectrum of musculoskeletal indications.ConclusionA dedicated extremity cone-beam CT scanner capable of imaging upper and lower extremities (including weight-bearing examinations) provides sufficient image quality and favorable dose characteristics to warrant further evaluation for clinical use.
| S-EPMC4263642 | biostudies-literature
Project description:Background & purposeFour dimensional Cone beam CT (CBCT) has many potential benefits for radiotherapy but suffers from poor image quality, long acquisition times, and/or long reconstruction times. In this work we present a fast iterative reconstruction algorithm for 4D reconstruction of fast acquisition cone beam CT, as well as a new method for temporal regularization and compare to state of the art methods for 4D CBCT.Materials & methodsRegularization parameters for the iterative algorithms were found automatically via computer optimization on 60 s acquisitions using the XCAT phantom. Nineteen lung cancer patients were scanned with 60 s arcs using the onboard image on a Varian trilogy linear accelerator. Images were reconstructed using an accelerated ordered subset algorithm. A frequency based temporal regularization algorithm was developed and compared to the McKinnon-Bates algorithm, 4D total variation and prior images compressed sensing (PICCS).ResultsAll reconstructions were completed in 60 s or less. The proposed method provided a structural similarity of 0.915, compared with 0.786 for the classic McKinnon-bates method. For the patient study, it provided fewer image artefacts than PICCS, and better spatial resolution than 4D TV.ConclusionFour dimensional iterative CBCT reconstruction was done in less than 60 s, demonstrating the clinical feasibility. The frequency based method outperformed 4D total variation and PICCS on the simulated data, and for patients allowed for tumor location based on 60 s acquisitions, even for slowly breathing patients. It should thus be suitable for routine clinical use.
| S-EPMC7807579 | biostudies-literature
Project description:We investigated emotion classification from brief video recordings from the GEMEP database wherein actors portrayed 18 emotions. Vocal features consisted of acoustic parameters related to frequency, intensity, spectral distribution, and durations. Facial features consisted of facial action units. We first performed a series of person-independent supervised classification experiments. Best performance (AUC = 0.88) was obtained by merging the output from the best unimodal vocal (Elastic Net, AUC = 0.82) and facial (Random Forest, AUC = 0.80) classifiers using a late fusion approach and the product rule method. All 18 emotions were recognized with above-chance recall, although recognition rates varied widely across emotions (e.g., high for amusement, anger, and disgust; and low for shame). Multimodal feature patterns for each emotion are described in terms of the vocal and facial features that contributed most to classifier performance. Next, a series of exploratory unsupervised classification experiments were performed to gain more insight into how emotion expressions are organized. Solutions from traditional clustering techniques were interpreted using decision trees in order to explore which features underlie clustering. Another approach utilized various dimensionality reduction techniques paired with inspection of data visualizations. Unsupervised methods did not cluster stimuli in terms of emotion categories, but several explanatory patterns were observed. Some could be interpreted in terms of valence and arousal, but actor and gender specific aspects also contributed to clustering. Identifying explanatory patterns holds great potential as a meta-heuristic when unsupervised methods are used in complex classification tasks.
| S-EPMC8725659 | biostudies-literature
Project description:ObjectivesThe aim of this study was to assess the consistency of therapy radiographers performing image registration using cone beam computed tomography (CBCT)-CT, magnetic resonance (MR)-CT, and MR-MR image guidance for cervix cancer radiotherapy and to assess that MR-based image guidance is not inferior to CBCT standard practice.Methods10 patients receiving cervix radiation therapy underwent daily CBCT guidance and magnetic resonance (MR) imaging weekly during treatment. Offline registration of each MR image, and corresponding CBCT, to planning CT was performed by five radiographers. MR images were also registered to the earliest MR interobserver variation was assessed using modified Bland-Altman analysis with clinically acceptable 95% limits of agreement (LoA) defined as ±5.0 mm.Results30 CBCT-CT, 30 MR-CT and 20 MR-MR registrations were performed by each observer. Registration variations between CBCT-CT and MR-CT were minor and both strategies resulted in 95% LoA over the clinical threshold in the anteroposterior direction (CBCT-CT ±5.8 mm, MR-CT ±5.4 mm). MR-MR registrations achieved a significantly improved 95% LoA in the anteroposterior direction (±4.3 mm). All strategies demonstrated similar results in lateral and longitudinal directions.ConclusionThe magnitude of interobserver variations between CBCT-CT and MR-CT were similar, confirming that MR-CT radiotherapy workflows are comparable to CBCT-CT image-guided radiotherapy. Our results suggest MR-MR radiotherapy workflows may be a superior registration strategy.Advances in knowledgeThis is the first publication quantifying interobserver registration of multimodality image registration strategies for cervix radical radiotherapy patients.
| S-EPMC7446016 | biostudies-literature
Project description:Background and purposeThe limited image quality in Cone Beam CT (CBCT) stemming primarily from scattered radiation hinders accurate CBCT based dose calculation in radiotherapy. We investigated the use of a stoichiometric calibration for dose calculation on CBCT images of lung cancer patients.Materials and methodsCBCT calibrations were performed with thorax scan protocols, using a phantom with approximately the diameter of an average human thorax and a central cavity simulating the thoracic cavity. Thus scatter conditions resembling those in clinical thorax CBCT scans were simulated. A published stoichiometric parametrization was used. A treatment plan was simulated on CBCT and CT scans of an anthropomorphic phantom, the dose distributions were calculated, and clinically relevant DVH parameters were compared. Twelve lung cancer patients had surveillance CT scans (s-CT) taken twice during their treatment course in addition to daily setup CBCTs. Dose calculations were performed on the s-CTs and the corresponding CBCTs taken on the same day, and DVH parameters were compared.ResultsEighty percent of CBCT DVH parameters found for the phantom were within ±1% of CT doses, and 98% were within ±3%. For patients, the median CT/CBCT dose difference was within ±2%, and 98% of DVH parameters were within ±4%. Minimum dose to the tumor was underestimated (median 1.9%) on CBCT, while maximum doses to most organs at risk were slightly overestimated.ConclusionDirect dose calculations on CBCTs of lung cancer patients were feasible within ∼4% accuracy using a simple calibration method, which is easily implemented in a clinical setting.
| S-EPMC7807659 | biostudies-literature
Project description:BackgroundMinimally invasive surgery (MIS) for evacuation of spontaneous intracerebral hemorrhage (ICH) has shown promise but there remains a need for intraoperative performance assessment considering the wide range of evacuation effectiveness. In this feasibility study, we analyzed the benefit of intraoperative 3-dimensional imaging during navigated endoscopy-assisted ICH evacuation by mechanical clot fragmentation and aspiration.Methods18 patients with superficial or deep supratentorial ICH underwent MIS for clot evacuation followed by intraoperative computerized tomography (iCT) or cone-beam CT (CBCT) imaging. Eligibility for MIS required (a) availability of intraoperative iCT or CBCT, (b) spontaneous lobar or deep ICH without vascular pathology, (c) a stable ICH volume (20-90 ml), (d) a reduced level of consciousness (GCS 5-14), and (e) a premorbid mRS ≤ 1. Demographic, clinical, and radiographic patient data were analyzed by two independent observers.ResultsNine female and 9 male patients with a median age of 76 years (42-85) presented with an ICH score of 3 (1-4), GCS of 10 (5-14) and ICH volume of 54 ± 26 ml. Clot fragmentation and aspiration was feasible in all cases and intraoperative imaging determined an overall evacuation rate of 80 ± 19% (residual hematoma volume: 13 ± 17 ml; p < 0.0001 vs. Pre-OP). Based on the intraoperative imaging results, 1/3rd of all patients underwent an immediate re-aspiration attempt. No patient experienced hemorrhagic complications or required conversion to open craniotomy. However, routine postoperative CT imaging revealed early hematoma re-expansion with an adjusted evacuation rate of 59 ± 30% (residual hematoma volume: 26 ± 37 ml; p < 0.001 vs. Pre-OP).ConclusionsRoutine utilization of iCT or CBCT imaging in MIS for ICH permits direct surgical performance assessment and the chance for immediate re-aspiration, which may optimize targeting of an ideal residual hematoma volume and reduce secondary revision rates.
| S-EPMC7593312 | biostudies-literature
Project description:In computer vision, reference datasets from simulation and real outdoor scenes have been highly successful in promoting algorithmic development in stereo reconstruction. Endoscopic stereo reconstruction for surgical scenes gives rise to specific problems, including the lack of clear corner features, highly specular surface properties and the presence of blood and smoke. These issues present difficulties for both stereo reconstruction itself and also for standardised dataset production. Previous datasets have been produced using computed tomography (CT) or structured light reconstruction on phantom or ex vivo models. We present a stereo-endoscopic reconstruction validation dataset based on cone-beam CT (SERV-CT). Two ex vivo small porcine full torso cadavers were placed within the view of the endoscope with both the endoscope and target anatomy visible in the CT scan. Subsequent orientation of the endoscope was manually aligned to match the stereoscopic view and benchmark disparities, depths and occlusions are calculated. The requirement of a CT scan limited the number of stereo pairs to 8 from each ex vivo sample. For the second sample an RGB surface was acquired to aid alignment of smooth, featureless surfaces. Repeated manual alignments showed an RMS disparity accuracy of around 2 pixels and a depth accuracy of about 2 mm. A simplified reference dataset is provided consisting of endoscope image pairs with corresponding calibration, disparities, depths and occlusions covering the majority of the endoscopic image and a range of tissue types, including smooth specular surfaces, as well as significant variation of depth. We assessed the performance of various stereo algorithms from online available repositories. There is a significant variation between algorithms, highlighting some of the challenges of surgical endoscopic images. The SERV-CT dataset provides an easy to use stereoscopic validation for surgical applications with smooth reference disparities and depths covering the majority of the endoscopic image. This complements existing resources well and we hope will aid the development of surgical endoscopic anatomical reconstruction algorithms.
| S-EPMC8961000 | biostudies-literature
Project description:OBJECTIVES::To determine the optimized kV setting for a narrow detector cone-beam CT (CBCT) unit. METHODS::Clinical (CL) and quantitative (QUANT) evaluations of image quality were performed using an anthropomorphic phantom. Technical (TECH) evaluation was performed with a polymethyl methacrylate phantom. Images were obtained using a PaX-i3D Green CBCT (Vatech, Hwaseong, Korea) device, with a large 21 × 19?and a medium 12 × 9?cm field of view (FOV), and high-dose (HD-ranging from 85 to 110 kV) and low-dose (LD-ranging from 75 to 95 kV) protocols, totaling four groups (21 × 19?cm HD, 21 × 19?cm LD, 12 × 9?cm HD, 12 × 9?cm LD). The radiation dose within each group was fixed by adapting the mA according to a predetermined dose-area product. For CL evaluation, three observers assessed images based on overall quality, sharpness, contrast, artefacts, and noise. For QUANT evaluation, mean gray value shift, % increase of standard deviation (SD), % of beam hardening and contrast-to-noise ratio (CNR) were calculated. For TECH evaluation, segmentation accuracy, CNR, metal artefact SD, metal object area, and sharpness were measured. Representative parameters were chosen for CL, QUANT, and TECH evaluations to determine the optimal kV based on biplot graphs. kV values of the same protocol were compared by the bootstrapping approach. The ones that had statistical differences with the best kV were considered as worse quality. RESULTS::Overall, kV values within the same group showed similar quality (p > 0.05), except for 110 kV in 21 × 19?cm HD and 85 kV in 12 × 9?cm HD of CL score; also 85, 90 kV in 21 × 19?cm HD and 75, 80 kV in 21 × 19?cm LD of QUANT score which were worse (p < 0.05). CONCLUSION::At a constant dose, low and high kV protocols yield acceptable image quality for a narrow-detector CBCT unit.
| S-EPMC6476352 | biostudies-literature
Project description:ObjectiveTo determine whether odontometric parameters using cone beam computed tomography (CBCT) would aid in sex estimation by assessing sexual dimorphism of odontometric parameters.Material and methodsThe focused question was whether there is sexual dimorphism in linear and volumetric odontometric parameters when assessed using CBCT. The preferred reporting items for systematic reviews and meta-analysis (PRISMA) guidelines were followed to conduct a systematic search until June 2022 in all major databases. Data were extracted regarding the population, size of the sample, age range, teeth analyzed, linear or volumetric measurements, accuracy, and conclusion. The quality of included studies was assessed using (Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool.ResultsOut of the 3761 studies identified, twenty-nine full-text articles were assessed for eligibility. Finally, twenty-three articles (4215 participants) that provided data on odontometrics using CBCT were included in this systematic review. The odontological sex estimation were assessed either linear measurements (n = 13) or volumetric measurements (n = 8) or both (n = 2). Canines were analysed in maximum number of reports (n = 14), followed by incisors (n = 11), molars(n = 10) and premolars(n = 6). Most of the reports (n = 18) confirmed the existence of sexual dimorphism in odontometric parameters when assessed using CBCT. No significant differences in odontometrics between the sexes were noted in some reports (n = 5). The accuracy of sex estimation was assessed in eight investigations, which ranged from 47.8 to 92.3%.ConclusionsOdontometrics of human permanent dentition using CBCT exhibit a certain degree of sexual dimorphism. Both linear and volumetric measurements of teeth can aid sex estimation.
| S-EPMC9990232 | biostudies-literature