Differentiating loss of consciousness causes through artificial intelligence-enabled decoding of functional connectivity.

Differential diagnosis of acute loss of consciousness (LOC) is crucial due to the need for different therapeutic strategies despite similar clinical presentations among etiologies such as nonconvulsive status epilepticus, metabolic encephalopathy, and benzodiazepine intoxication. While altered functional connectivity (FC) plays a pivotal role in the pathophysiology of LOC, there has been a lack of efforts to develop differential diagnosis artificial intelligence (AI) models that feature the distinctive FC change patterns specific to each LOC cause. Three approaches were applied for extracting features for the AI models: three-dimensional FC adjacency matrices, vectorized FC values, and graph theoretical measurements. Deep learning using convolutional neural networks (CNN) and various machine learning algorithms were implemented to compare classification accuracy using electroencephalography (EEG) data with different epoch sizes. The CNN model using FC adjacency matrices achieved the highest accuracy with an AUC of 0.905, with 20-s epoch data being optimal for classifying the different LOC causes. The high accuracy of the CNN model was maintained in a prospective cohort. Key distinguishing features among the LOC causes were found in the delta and theta brain wave bands. This research advances the understanding of LOC's underlying mechanisms and shows promise for enhancing diagnosis and treatment selection. Moreover, the AI models can provide accurate LOC differentiation with a relatively small amount of EEG data in 20-s epochs, which may be clinically useful.

Deep Convolutional Neural Network-based Software Improves Radiologist Detection of Malignant Lung Nodules on Chest Radiographs.

Background Multicenter studies are required to validate the added benefit of using deep convolutional neural network (DCNN) software for detecting malignant pulmonary nodules on chest radiographs. Purpose To compare the performance of radiologists in detecting malignant pulmonary nodules on chest radiographs when assisted by deep learning-based DCNN software with that of radiologists or DCNN software alone in a multicenter setting. Materials and Methods Investigators at four medical centers retrospectively identified 600 lung cancer-containing chest radiographs and 200 normal chest radiographs. Each radiograph with a lung cancer had at least one malignant nodule confirmed by CT and pathologic examination. Twelve radiologists from the four centers independently analyzed the chest radiographs and marked regions of interest. Commercially available deep learning-based computer-aided detection software separately trained, tested, and validated with 19 330 radiographs was used to find suspicious nodules. The radiologists then reviewed the images with the assistance of DCNN software. The sensitivity and number of false-positive findings per image of DCNN software, radiologists alone, and radiologists with the use of DCNN software were analyzed by using logistic regression and Poisson regression. Results The average sensitivity of radiologists improved (from 65.1% [1375 of 2112; 95% confidence interval {CI}: 62.0%, 68.1%] to 70.3% [1484 of 2112; 95% CI: 67.2%, 73.1%], P < .001) and the number of false-positive findings per radiograph declined (from 0.2 [488 of 2400; 95% CI: 0.18, 0.22] to 0.18 [422 of 2400; 95% CI: 0.16, 0.2], P < .001) when the radiologists re-reviewed radiographs with the DCNN software. For the 12 radiologists in this study, 104 of 2400 radiographs were positively changed (from false-negative to true-positive or from false-positive to true-negative) using the DCNN, while 56 of 2400 radiographs were changed negatively. Conclusion Radiologists had better performance with deep convolutional network software for the detection of malignant pulmonary nodules on chest radiographs than without. © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Jacobson in this issue.

Artificial intelligence-assisted interpretation of bone age radiographs improves accuracy and decreases variability.

OBJECTIVE: Radiographic bone age assessment (BAA) is used in the evaluation of pediatric endocrine and metabolic disorders. We previously developed an automated artificial intelligence (AI) deep learning algorithm to perform BAA using convolutional neural networks. We compared the BAA performance of a cohort of pediatric radiologists with and without AI assistance. MATERIALS AND METHODS: Six board-certified, subspecialty trained pediatric radiologists interpreted 280 age- and gender-matched bone age radiographs ranging from 5 to 18 years. Three of those radiologists then performed BAA with AI assistance. Bone age accuracy and root mean squared error (RMSE) were used as measures of accuracy. Intraclass correlation coefficient evaluated inter-rater variation. RESULTS: AI BAA accuracy was 68.2% overall and 98.6% within 1 year, and the mean six-reader cohort accuracy was 63.6 and 97.4% within 1 year. AI RMSE was 0.601 years, while mean single-reader RMSE was 0.661 years. Pooled RMSE decreased from 0.661 to 0.508 years, all individually decreasing with AI assistance. ICC without AI was 0.9914 and with AI was 0.9951. CONCLUSIONS: AI improves radiologist's bone age assessment by increasing accuracy and decreasing variability and RMSE. The utilization of AI by radiologists improves performance compared to AI alone, a radiologist alone, or a pooled cohort of experts. This suggests that AI may optimally be utilized as an adjunct to radiologist interpretation of imaging studies to improve performance.

Beyond Human Perception: Sexual Dimorphism in Hand and Wrist Radiographs Is Discernible by a Deep Learning Model.

Despite the well-established impact of sex and sex hormones on bone structure and density, there has been limited description of sexual dimorphism in the hand and wrist in the literature. We developed a deep convolutional neural network (CNN) model to predict sex based on hand radiographs of children and adults aged between 5 and 70 years. Of the 1531 radiographs tested, the algorithm predicted sex correctly in 95.9% (κ = 0.92) of the cases. Two human radiologists achieved 58% (κ = 0.15) and 46% (κ = - 0.07) accuracy. The class activation maps (CAM) showed that the model mostly focused on the 2nd and 3rd metacarpal base or thumb sesamoid in women, and distal radioulnar joint, distal radial physis and epiphysis, or 3rd metacarpophalangeal joint in men. The radiologists reviewed 70 cases (35 females and 35 males) labeled with sex along with heat maps generated by CAM, but they could not find any patterns that distinguish the two sexes. A small sample of patients (n = 44) with sexual developmental disorders or transgender identity was selected for a preliminary exploration of application of the model. The model prediction agreed with phenotypic sex in only 77.8% (κ = 0.54) of these cases. To the best of our knowledge, this is the first study that demonstrated a machine learning model to perform a task in which human experts could not fulfill.

Current Applications and Future Impact of Machine Learning in Radiology.

Recent advances and future perspectives of machine learning techniques offer promising applications in medical imaging. Machine learning has the potential to improve different steps of the radiology workflow including order scheduling and triage, clinical decision support systems, detection and interpretation of findings, postprocessing and dose estimation, examination quality control, and radiology reporting. In this article, the authors review examples of current applications of machine learning and artificial intelligence techniques in diagnostic radiology. In addition, the future impact and natural extension of these techniques in radiology practice are discussed.

Quantifying the effect of slice thickness, intravenous contrast and tube current on muscle segmentation: Implications for body composition analysis.

OBJECTIVES: To quantify the effect of IV contrast, tube current and slice thickness on skeletal muscle cross-sectional area (CSA) and density (SMD) on routine CT. METHODS: CSA and SMD were computed on 216 axial CT images obtained at the L3 level in 72 patients with variations in IV contrast, slice thickness and tube current. Intra-patient mean difference (MD), 95 % CI and limits of agreement were calculated using the Bland-Altman approach. Inter- and intra-analyst agreement was evaluated. RESULTS: IV contrast significantly increased CSA by 1.88 % (MD 2.33 cm2; 95 % CI 1.76-2.89) and SMD by 5.99 % (p<0.0001). Five mm slice thickness significantly increased mean CSA by 1.11 % compared to 2 mm images (1.32 cm2; 0.78-1.85) and significantly decreased SMD by 11.64 % (p<0.0001). Low tube current significantly decreased mean CSA by 4.79 % (6.44 cm2; 3.78-9.10) and significantly increased SMD by 46.46 % (p<0.0001). Inter- and intra-analyst agreement was excellent. CONCLUSIONS: IV contrast, slice thickness and tube current significantly affect CSA and SMD. Investigators designing and analysing clinical trials using CT for body composition analysis should report CT acquisition parameters and consider the effect of slice thickness, IV contrast and tube current on myometric data. KEY POINTS: • Intravenous contrast, slice thickness and tube current significantly affect myometric data. • Image acquisition parameter variations may obscure intrapatient muscle differences on serial measurements. • Investigators using CT for body composition analysis should report CT acquisition parameters.

A Deep-Learning System for Fully-Automated Peripherally Inserted Central Catheter (PICC) Tip Detection.

A peripherally inserted central catheter (PICC) is a thin catheter that is inserted via arm veins and threaded near the heart, providing intravenous access. The final catheter tip position is always confirmed on a chest radiograph (CXR) immediately after insertion since malpositioned PICCs can cause potentially life-threatening complications. Although radiologists interpret PICC tip location with high accuracy, delays in interpretation can be significant. In this study, we proposed a fully-automated, deep-learning system with a cascading segmentation AI system containing two fully convolutional neural networks for detecting a PICC line and its tip location. A preprocessing module performed image quality and dimension normalization, and a post-processing module found the PICC tip accurately by pruning false positives. Our best model, trained on 400 training cases and selectively tuned on 50 validation cases, obtained absolute distances from ground truth with a mean of 3.10 mm, a standard deviation of 2.03 mm, and a root mean squares error (RMSE) of 3.71 mm on 150 held-out test cases. This system could help speed confirmation of PICC position and further be generalized to include other types of vascular access and therapeutic support devices.

Iterative Image Reconstruction Improves the Accuracy of Automated Plaque Burden Assessment in Coronary CT Angiography: A Comparison With Intravascular Ultrasound.

OBJECTIVE: The purpose of this study was to determine whether use of iterative image reconstruction algorithms improves the accuracy of coronary CT angiography (CCTA) compared with intravascular ultrasound (IVUS) in semiautomated plaque burden assessment. MATERIALS AND METHODS: CCTA and IVUS images of seven coronary arteries were acquired ex vivo. CT images were reconstructed with filtered back projection (FBP) and adaptive statistical (ASIR) and model-based (MBIR) iterative reconstruction algorithms. Cross-sectional images of the arteries were coregistered between CCTA and IVUS in 1-mm increments. In CCTA, fully automated (without manual corrections) and semiautomated (allowing manual corrections of vessel wall boundaries) plaque burden assessments were performed for each of the reconstruction algorithms with commercially available software. In IVUS, plaque burden was measured manually. Agreement between CCTA and IVUS was determined with Pearson correlation. RESULTS: A total of 173 corresponding cross sections were included. The mean plaque burden measured with IVUS was 63.39% ± 10.63%. With CCTA and the fully automated technique, it was 54.90% ± 11.70% with FBP, 53.34% ± 13.11% with ASIR, and 55.35% ± 12.22% with MBIR. With CCTA and the semiautomated technique mean plaque burden was 54.90% ± 11.76%, 53.40% ± 12.85%, 57.09% ± 11.05%. Manual correction of the semiautomated assessments was performed in 39% of all cross sections and improved plaque burden correlation with the IVUS assessment independently of reconstruction algorithm (p < 0.0001). Furthermore, MBIR was superior to FBP and ASIR independently of assessment method (semiautomated, r = 0.59 for FBP, r = 0.52 for ASIR, r = 0.78 for MBIR, all p < 0.001; fully automated, r = 0.40 for FBP, r = 0.37 for ASIR, r = 0.53 for MBIR, all p < 0.001). CONCLUSION: For the quantification of plaque burden with CCTA, MBIR led to better correlation with IVUS than did traditional reconstruction algorithms such as FBP, independently of the use of a fully automated or semiautomated assessment approach. The highest accuracy for quantifying plaque burden with CCTA can be achieved by using MBIR data with semiautomated assessment.

Fully Automated Deep Learning System for Bone Age Assessment.

Skeletal maturity progresses through discrete phases, a fact that is used routinely in pediatrics where bone age assessments (BAAs) are compared to chronological age in the evaluation of endocrine and metabolic disorders. While central to many disease evaluations, little has changed to improve the tedious process since its introduction in 1950. In this study, we propose a fully automated deep learning pipeline to segment a region of interest, standardize and preprocess input radiographs, and perform BAA. Our models use an ImageNet pretrained, fine-tuned convolutional neural network (CNN) to achieve 57.32 and 61.40% accuracies for the female and male cohorts on our held-out test images. Female test radiographs were assigned a BAA within 1 year 90.39% and within 2 years 98.11% of the time. Male test radiographs were assigned 94.18% within 1 year and 99.00% within 2 years. Using the input occlusion method, attention maps were created which reveal what features the trained model uses to perform BAA. These correspond to what human experts look at when manually performing BAA. Finally, the fully automated BAA system was deployed in the clinical environment as a decision supporting system for more accurate and efficient BAAs at much faster interpretation time (<2 s) than the conventional method.

Pixel-Level Deep Segmentation: Artificial Intelligence Quantifies Muscle on Computed Tomography for Body Morphometric Analysis.

Pretreatment risk stratification is key for personalized medicine. While many physicians rely on an "eyeball test" to assess whether patients will tolerate major surgery or chemotherapy, "eyeballing" is inherently subjective and difficult to quantify. The concept of morphometric age derived from cross-sectional imaging has been found to correlate well with outcomes such as length of stay, morbidity, and mortality. However, the determination of the morphometric age is time intensive and requires highly trained experts. In this study, we propose a fully automated deep learning system for the segmentation of skeletal muscle cross-sectional area (CSA) on an axial computed tomography image taken at the third lumbar vertebra. We utilized a fully automated deep segmentation model derived from an extended implementation of a fully convolutional network with weight initialization of an ImageNet pre-trained model, followed by post processing to eliminate intramuscular fat for a more accurate analysis. This experiment was conducted by varying window level (WL), window width (WW), and bit resolutions in order to better understand the effects of the parameters on the model performance. Our best model, fine-tuned on 250 training images and ground truth labels, achieves 0.93 ± 0.02 Dice similarity coefficient (DSC) and 3.68 ± 2.29% difference between predicted and ground truth muscle CSA on 150 held-out test cases. Ultimately, the fully automated segmentation system can be embedded into the clinical environment to accelerate the quantification of muscle and expanded to volume analysis of 3D datasets.

Ultralow-Dose Abdominal Computed Tomography: Comparison of 2 Iterative Reconstruction Techniques in a Prospective Clinical Study.

PURPOSE: To assess lesion detection and image quality of ultralow-dose (ULD) abdominal computed tomography (CT) reconstructed with filtered back projection (FBP) and 2 iterative reconstruction techniques: hybrid-based iDose, and image-based SafeCT. MATERIALS AND METHODS: In this institutional review board-approved ongoing prospective clinical study, 41 adult patients provided written informed consent for an additional ULD abdominal CT examination immediately after standard dose (SD) CT exam on a 256-slice multidetector computed tomography (iCT, Philips-Healthcare). The SD examination (size-specific dose estimate, 10 ± 3 mGy) was performed at 120 kV with automatic exposure control, and reconstructed with FBP. The ULD examination (1.5 ± 0.4 mGy) was performed at 120 kV and fixed tube current of 17 to 20 mAs/slice to achieve ULD radiation dose, with the rest of the scan parameters same as SD examination. The ULD data were reconstructed with (a) FBP, (b) iDose, and (c) SafeCT. Lesions were detected on ULD FBP series and compared to SD FBP "reference-standard" series. True lesions, pseudolesions, and missed lesions were recorded. Four abdominal radiologists independently blindly performed subjective image quality. Objective image quality included image noise calculation and noise spectral density plots. RESULTS: All true lesions (n, 52: liver metastases, renal cysts, diverticulosis) in SD FBP images were detected in ULD images. Although there were no missed or pseudolesions on ULD iDose and ULD SafeCT images, appearance of small low-contrast hepatic lesions was suboptimal. The ULD FBP images were unacceptable across all patients for both lesion detection and image quality. In patients with a body mass index (BMI) of 25 kg/m or less, ULD iDose and ULD SafeCT images were acceptable for image quality that was close to SD FBP for both normal and abnormal abdominal and pelvic structures. With increasing BMI, the image quality of ULD images was deemed unacceptable due to photo starvation. Evaluation of kidney stones with ULD iDose/SafeCT images was found acceptable regardless of patient size. Image noise levels were significantly lower in ULD iDose and ULD SafeCT images compared to ULD FBP (P < 0.01). CONCLUSIONS: Preliminary results show that ULD abdominal CT reconstructed with iterative reconstruction techniques is achievable in smaller patients (BMI ≤ 25 kg/m) but remains a challenge for overweight to obese patients. Lesion detection is similar in full-dose SD FBP and ULD iDose/SafeCT images, with suboptimal visibility of low-contrast lesions in ULD images.

Assessment of Filtered Back Projection, Adaptive Statistical, and Model-Based Iterative Reconstruction for Reduced Dose Abdominal Computed Tomography.

PURPOSE: To compare standard of care and reduced dose (RD) abdominal computed tomography (CT) images reconstructed with filtered back projection (FBP), adaptive statistical iterative reconstruction (ASIR), model-based iterative reconstruction (MBIR) techniques. MATERIALS AND METHODS: In an Institutional Review Board-approved, prospective clinical study, 28 patients (mean age 59 ± 13 years ), undergoing clinically indicated routine abdominal CT on a 64-channel multi-detector CT scanner, gave written informed consent for acquisition of an additional RD (<1 milli-Sievert) abdomen CT series. Sinogram data of RD series were reconstructed with FBP, ASIR, and MBIR and compared with FBP images of standard dose abdomen CT. Two radiologists performed randomized, independent, and blinded comparison for lesion detection, lesion margin, visibility of normal structures, and diagnostic confidence. RESULTS: Mean CT dose index volume was 10 ± 3.4 mGy and 1.3 ± 0.3 mGy for standard and RD CT, respectively. There were 73 "true positive" lesions detected on standard of care CT. Nine lesions (<8 mm in size) were missed on RD abdominal CT images which included liver lesions, liver cysts, kidney cysts, and paracolonic abscess. These lesions were missed regardless of patient size and types of iterative reconstruction techniques used for reconstruction of RD data sets. The visibility of lesion margin was suboptimal in (23/28) patients with RD FBP, (15/28) patients with RD ASIR, and (14/28) patients with RD MBIR compared to standard of care FBP images (P < 0.001). Diagnostic confidence for the assessment of lesions on RD images was suboptimal in most patients regardless of iterative reconstruction techniques. CONCLUSIONS: Clinically significant lesions (< 8 mm) can be missed on abdominal CT examinations acquired at a CT dose index volume of 1.3 mGy regardless of patients' size and reconstruction techniques (FBP, ASIR, and MBIR).

Dose reduction for chest CT: comparison of two iterative reconstruction techniques.

BACKGROUND: Lowering radiation dose in computed tomography (CT) scan results in low quality noisy images. Iterative reconstruction techniques are used currently to lower image noise and improve the quality of images. PURPOSE: To evaluate lesion detection and diagnostic acceptability of chest CT images acquired at CTDIvol of 1.8 mGy and processed with two different iterative reconstruction techniques. MATERIAL AND METHODS: Twenty-two patients (mean age, 60 ± 14 years; men, 13; women, 9; body mass index, 27.4 ± 6.5 kg/m(2)) gave informed consent for acquisition of low dose (LD) series in addition to the standard dose (SD) chest CT on a 128 - multidetector CT (MDCT). LD images were reconstructed with SafeCT C4, L1, and L2 settings, and Safire S1, S2, and S3 settings. Three thoracic radiologists assessed LD image series (S1, S2, S3, C4, L1, and L2) for lesion detection and comparison of lesion margin, visibility of normal structures, and diagnostic confidence with SD chest CT. Inter-observer agreement (kappa) was calculated. RESULTS: Average CTDIvol was 6.4 ± 2.7 mGy and 1.8 ± 0.2 mGy for SD and LD series, respectively. No additional lesion was found in SD as compared to LD images. Visibility of ground-glass opacities and lesion margins, as well as normal structures visibility were not affected on LD. CT image visibility of major fissure and pericardium was not optimal in some cases (n = 5). Objective image noise in some low dose images processed with SafeCT and Safire was similar to SD images (P value > 0.5). CONCLUSION: Routine LD chest CT reconstructed with iterative reconstruction technique can provide similar diagnostic information in terms of lesion detection, margin, and diagnostic confidence as compared to SD, regardless of the iterative reconstruction settings.

Crystal analyser-based X-ray phase contrast imaging in the dark field: implementation and evaluation using excised tissue specimens.

OBJECTIVES: We demonstrate the soft tissue discrimination capability of X-ray dark-field imaging (XDFI) using a variety of human tissue specimens. METHODS: The experimental setup for XDFI comprises an X-ray source, an asymmetrically cut Bragg-type monochromator-collimator (MC), a Laue-case angle analyser (LAA) and a CCD camera. The specimen is placed between the MC and the LAA. For the light source, we used the beamline BL14C on a 2.5-GeV storage ring in the KEK Photon Factory, Tsukuba, Japan. RESULTS: In the eye specimen, phase contrast images from XDFI were able to discriminate soft-tissue structures, such as the iris, separated by aqueous humour on both sides, which have nearly equal absorption. Superiority of XDFI in imaging soft tissue was further demonstrated with a diseased iliac artery containing atherosclerotic plaque and breast samples with benign and malignant tumours. XDFI on breast tumours discriminated between the normal and diseased terminal duct lobular unit and between invasive and in-situ cancer. CONCLUSIONS: X-ray phase, as detected by XDFI, has superior contrast over absorption for soft tissue processes such as atherosclerotic plaque and breast cancer. KEY POINTS: • X-ray dark field imaging (XDFI) can dramatically increase sensitivity of phase detection. • XDFI can provide enhanced soft tissue discrimination. • With XDFI, abnormal anatomy can be visualised with high spatial/contrast resolution.

Submillisievert chest CT with filtered back projection and iterative reconstruction techniques.

OBJECTIVE: The purpose of this study was to compare submillisievert chest CT images reconstructed with filtered back projection (FBP), SafeCT, adaptive statistical iterative reconstruction (ASIR), and model-based iterative reconstruction (MBIR) with standard of care FBP images. SUBJECTS AND METHODS: Fifty patients (33 men and 17 women; mean age [± SD], 62 ± 10 years) undergoing routine chest CT gave written informed consent for acquisition of an additional submillisievert chest CT series with reduced tube current but identical scanning length as standard of care chest CT. Sinogram data of the submillisievert series were reconstructed with FBP, SafeCT, ASIR, and MBIR and compared with FBP images at standard-dose chest CT (n = 8 × 50 = 400 series). Two thoracic radiologists performed independent comparison for visualization of lesion margin, visibility of small structures, and diagnostic acceptability. Objective noise measurements and noise spectral density were obtained. RESULTS: Of 287 detected lesions, 162 were less than 1-cm noncalcified nodules. Lesion margins were well seen on all submillisievert reconstruction images except MBIR, on which they were poorly visualized. Likewise, only submillisievert MBIR images were suboptimal for visibility of normal structures, such as pulmonary vessels in the outer 2 cm of the lung, interlobular fissures, and subsegmental bronchial walls. MBIR had the lowest image noise compared with other techniques. CONCLUSION: FBP, SafeCT, ASIR, and MBIR can enable optimal lesion evaluation on chest CT acquired at a volume CT dose index of 2 mGy. However, all submillisievert reconstruction techniques were suboptimal for visualization of mediastinal structures. Submillisievert MBIR images were suboptimal for visibility of normal lung structures despite showing lower image noise.

Role of compressive sensing technique in dose reduction for chest computed tomography: a prospective blinded clinical study.

PURPOSE: The purpose of this study was to assess pulmonary lesion detection, diagnostic confidence, and noise reduction in sparse-sampled (SpS) computed tomographic (CT) data of submillisievert (SubmSv) chest CT reconstructed with iterative reconstruction technique (IRT). MATERIALS AND METHODS: This Human Insurance Portability and Accountability-compliant, institutional review board-approved prospective study was performed using SpS-SubmSv IRT chest CT in 10 non-obese patients (body-mass index, 21-35 kg/m; age range, 26-90 years). Written informed consent was obtained. The patients were scanned at standard-dose CT (mean [SD] volumetric CT dose index, 6 [0.9] mGy; mean [SD] dose-length product, 208 ± 44 mGy·cm; and mean [SD] effective dose, 3 [0.6] mSv) and at SubmSv dose (1.8 [0.2] mGy, 67 [2] mGy·cm, 0.9 [0.03] mSv, respectively) on a Philips 128-slice CT scanner with double z-sampling. Sparse angular sampling data were reconstructed using 25% of the angular projections from the SubmSv sinogram to reduce the number of views and radiation dose by approximately 4-fold. Hence, the patients were scanned and then, simulation-based sparse sampling was performed with a resultant dose hypothetical SpS scan estimated mathematically (0.2 mSv). From each patient data, 3 digital imaging and communications in medicine series were generated: SpS-SubmSv with IRT, fully sampled SubmSv filtered back projection (FBP), and fully sampled standard-dose FBP (SD-FBP). Two radiologists independently assessed these image series for detection of lung lesions, visibility of small structures, and diagnostic acceptability. Objective noise was measured in the thoracic aorta, and noise spectral density was obtained for SpS-SubmSv IRT, SubmSv-FBP, and SD-FBP. RESULTS: The SpS-SubmSv IRT resulted in 75% (0.2/0.9 mSv) and 92% (0.2/2.9 mSv) dose reduction, when compared with the fully sampled SubmSv-FBP and SD-FBP, respectively. Images of SpS-SubmSv displayed all 46 lesions (most <1 cm, 30 lung nodules, 7 ground glass opacities, 9 emphysema) seen on the SubmSv-FBP and SD-FBP data sets. Lesion margins with sparse-sampled data were deemed acceptable compared with both SubmSv-FBP and SD-FBP. Overall diagnostic confidence was maintained with SpS-SubmSv IRT despite the presence of minor pixilation artifacts in 3 of 10 cases. The SpS-SubmSv IRT showed 63% and 38% noise reduction when compared with SubmSv-FBP (P < 0.0001) and SD-FBP (P < 0.01), respectively, with no significant change in Hounsfield unit values (P > 0.05). Noise-spectral density showed that SpS-SubmSv IRT gives a linear decrease over frequency in the semilog plot and an exponential decrease of noise power over frequency compared with SubmSv-FBP and SD-FBP. CONCLUSIONS: More than 90% dose reduction could be achieved with one-fourth sparse-sampled and SubmSv chest CT examination when reconstructed with IRT. Chest CT dose at one fourth of a millisievert with SpS is possible with optimal lesion detection and diagnostic confidence for the evaluation of pulmonary findings.

Computed tomography (CT) of the chest at less than 1 mSv: an ongoing prospective clinical trial of chest CT at submillisievert radiation doses with iterative model image reconstruction and iDose4 technique.

PURPOSE: To assess lesion detection and diagnostic confidence of computed tomography (CT) of the chest performed at less than 1 mSv with 2 iterative reconstruction (IR) techniques. MATERIALS AND METHODS: Ten patients gave written informed consent for the acquisitions of images at submillisievert dose (0.9 mSv), in addition to clinical standard-dose (SD) chest CT (2.9 mSv). Submillisievert images were reconstructed with iDose and iterative model reconstruction (IMR). Two radiologists assessed lesion detection, margins, diagnostic confidence, and visibility of small structures. Objective noise and noise spectral density were measured. RESULTS: Lesion detection was identical for standard-dose filtered back projection (FBP), submSv iDose, and submSv IMR. Lesion margins were better seen for 30% of detected lung lesions with submSv IMR compared to standard-dose FBP and submSv iDose (P < 0.05). Visibility of abdominal structures, and diagnostic confidence with submSv iDose and submSv IMR were similar to standard-dose FBP. There was 21% to 64% noise reduction with submSv IMR and 1% to 15% higher noise with iDose compared to standard-dose FBP (P < 0.0001). CONCLUSIONS: Submillisievert IMR improves delineation of lesion margins compared to standard-dose FBP and submSv iDose.

Preliminary results: prospective clinical study to assess image-based iterative reconstruction for abdominal computed tomography acquired at 2 radiation dose levels.

OBJECTIVE: The objective of this study was to compare image quality for abdominal computed tomographic (CT) images acquired at 200 and 50 mA s and reconstructed with image-based iterative reconstruction. MATERIALS AND METHODS: In this institutional review board-approved prospective study, 22 patients (mean [SD] age, 64.3 [14.4] years; male-female ratio, 12:10) gave informed consent for acquisition of additional abdominal CT images on 64-slice multi-detector CT (MDCT) (Siemens Definition Flash). Standard-dose images were acquired at 200 quality reference mA s, whereas low-dose images were acquired at 50 mA s (all series: 120 kV; 5-mm section thickness; pitch, 0.9:1). The low-dose images were reconstructed with a nonlinear 3-dimensional iterative image reconstruction (3D-IIR) (SafeCT; MedicVision, Tirat Carmel, Israel) (4 settings, namely, A1, A2, A3, and A4) and were assessed by 3 abdominal radiologists for lesion detection, image noise, and visibility of small structures. CATPHAN 500 was scanned at the respective doses to obtain noise spectral density and modulation transfer function. RESULTS: Subjective image noise was unacceptable at 50-mA s filtered back projection and improved to average in 50-mA s A1 and minimal or no noise in 50-mA s A4. However, the visibility of small structures was similar to standard-dose filtered back projection images on 50-mA s A2. Objective image noise was reduced to 66% for the 50-mA s 3D-IIR images (9.08 [2.3]/26.75 [6.8]). The modulation transfer function curve demonstrated resolution improvement in the low-dose images with the 3D-IIR technique, whereas the noise spectral density curve confirmed noise suppression in the 50-mA s 3D-IIR images. CONCLUSIONS: Three-dimensional iterative image reconstruction helps to lower image noise without affecting the visibility of small structures at "moderate" settings. Diagnostically acceptable abdominal CT examinations can be acquired at 75% lower-radiation dose with the help of the image-based iterative reconstruction technique.