FSM-DDTR: End-to-end feedback strategy for multi-objective De Novo drug design using transformers.

The design of compounds that target specific biological functions with relevant selectivity is critical in the context of drug discovery, especially due to the polypharmacological nature of most existing drug molecules. In recent years, in silico-based methods combined with deep learning have shown promising results in the de novo drug design challenge, leading to potential leads for biologically interesting targets. However, several of these methods overlook the importance of certain properties, such as validity rate and target selectivity, or simplify the generative process by neglecting the multi-objective nature of the pharmacological space. In this study, we propose a multi-objective Transformer-based architecture to generate drug candidates with desired molecular properties and increased selectivity toward a specific biological target. The framework consists of a Transformer-Decoder Generator that generates novel and valid compounds in the SMILES format notation, a Transformer-Encoder Predictor that estimates the binding affinity toward the biological target, and a feedback loop combined with a multi-objective optimization strategy to rank the generated molecules and condition the generating distribution around the targeted properties. The results demonstrate that the proposed architecture can generate novel and synthesizable small compounds with desired pharmacological properties toward a biologically relevant target. The unbiased Transformer-based Generator achieved superior performance in the novelty rate (97.38%) and comparable performance in terms of internal diversity, uniqueness, and validity against state-of-the-art baselines. The optimization of the unbiased Transformer-based Generator resulted in the generation of molecules exhibiting high binding affinity toward the Adenosine A2A Receptor (AA2AR) and possessing desirable physicochemical properties, where 99.36% of the generated molecules follow Lipinski's rule of five. Furthermore, the implementation of a feedback strategy, in conjunction with a multi-objective algorithm, effectively shifted the distribution of the generated molecules toward optimal values of molecular weight, molecular lipophilicity, topological polar surface area, synthetic accessibility score, and quantitative estimate of drug-likeness, without the necessity of prior training sets comprising molecules endowed with pharmacological properties of interest. Overall, this research study validates the applicability of a Transformer-based architecture in the context of drug design, capable of exploring the vast chemical representation space to generate novel molecules with improved pharmacological properties and target selectivity. The data and source code used in this study are available at: https://github.com/larngroup/FSM-DDTR.

DTITR: End-to-end drug-target binding affinity prediction with transformers.

The accurate identification of Drug-Target Interactions (DTIs) remains a critical turning point in drug discovery and understanding of the binding process. Despite recent advances in computational solutions to overcome the challenges of in vitro and in vivo experiments, most of the proposed in silico-based methods still focus on binary classification, overlooking the importance of characterizing DTIs with unbiased binding strength values to properly distinguish primary interactions from those with off-targets. Moreover, several of these methods usually simplify the entire interaction mechanism, neglecting the joint contribution of the individual units of each binding component and the interacting substructures involved, and have yet to focus on more explainable and interpretable architectures. In this study, we propose an end-to-end Transformer-based architecture for predicting drug-target binding affinity (DTA) using 1D raw sequential and structural data to represent the proteins and compounds. This architecture exploits self-attention layers to capture the biological and chemical context of the proteins and compounds, respectively, and cross-attention layers to exchange information and capture the pharmacological context of the DTIs. The results show that the proposed architecture is effective in predicting DTA, achieving superior performance in both correctly predicting the value of interaction strength and being able to correctly discriminate the rank order of binding strength compared to state-of-the-art baselines. The combination of multiple Transformer-Encoders was found to result in robust and discriminative aggregate representations of the proteins and compounds for binding affinity prediction, in which the addition of a Cross-Attention Transformer-Encoder was identified as an important block for improving the discriminative power of these representations. Overall, this research study validates the applicability of an end-to-end Transformer-based architecture in the context of drug discovery, capable of self-providing different levels of potential DTI and prediction understanding due to the nature of the attention blocks. The data and source code used in this study are available at: https://github.com/larngroup/DTITR.

Designing optimized drug candidates with Generative Adversarial Network.

Drug design is an important area of study for pharmaceutical businesses. However, low efficacy, off-target delivery, time consumption, and high cost are challenges and can create barriers that impact this process. Deep Learning models are emerging as a promising solution to perform de novo drug design, i.e., to generate drug-like molecules tailored to specific needs. However, stereochemistry was not explicitly considered in the generated molecules, which is inevitable in targeted-oriented molecules. This paper proposes a framework based on Feedback Generative Adversarial Network (GAN) that includes optimization strategy by incorporating Encoder-Decoder, GAN, and Predictor deep models interconnected with a feedback loop. The Encoder-Decoder converts the string notations of molecules into latent space vectors, effectively creating a new type of molecular representation. At the same time, the GAN can learn and replicate the training data distribution and, therefore, generate new compounds. The feedback loop is designed to incorporate and evaluate the generated molecules according to the multiobjective desired property at every epoch of training to ensure a steady shift of the generated distribution towards the space of the targeted properties. Moreover, to develop a more precise set of molecules, we also incorporate a multiobjective optimization selection technique based on a non-dominated sorting genetic algorithm. The results demonstrate that the proposed framework can generate realistic, novel molecules that span the chemical space. The proposed Encoder-Decoder model correctly reconstructs 99% of the datasets, including stereochemical information. The model's ability to find uncharted regions of the chemical space was successfully shown by optimizing the unbiased GAN to generate molecules with a high binding affinity to the Kappa Opioid and Adenosine [Formula: see text] receptor. Furthermore, the generated compounds exhibit high internal and external diversity levels 0.88 and 0.94, respectively, and uniqueness.

Explainable deep drug-target representations for binding affinity prediction.

BACKGROUND: Several computational advances have been achieved in the drug discovery field, promoting the identification of novel drug-target interactions and new leads. However, most of these methodologies have been overlooking the importance of providing explanations to the decision-making process of deep learning architectures. In this research study, we explore the reliability of convolutional neural networks (CNNs) at identifying relevant regions for binding, specifically binding sites and motifs, and the significance of the deep representations extracted by providing explanations to the model's decisions based on the identification of the input regions that contributed the most to the prediction. We make use of an end-to-end deep learning architecture to predict binding affinity, where CNNs are exploited in their capacity to automatically identify and extract discriminating deep representations from 1D sequential and structural data. RESULTS: The results demonstrate the effectiveness of the deep representations extracted from CNNs in the prediction of drug-target interactions. CNNs were found to identify and extract features from regions relevant for the interaction, where the weight associated with these spots was in the range of those with the highest positive influence given by the CNNs in the prediction. The end-to-end deep learning model achieved the highest performance both in the prediction of the binding affinity and on the ability to correctly distinguish the interaction strength rank order when compared to baseline approaches. CONCLUSIONS: This research study validates the potential applicability of an end-to-end deep learning architecture in the context of drug discovery beyond the confined space of proteins and ligands with determined 3D structure. Furthermore, it shows the reliability of the deep representations extracted from the CNNs by providing explainability to the decision-making process.

Drug-Target Interaction Prediction: End-to-End Deep Learning Approach.

The discovery of potential Drug-Target Interactions (DTIs) is a determining step in the drug discovery and repositioning process, as the effectiveness of the currently available antibiotic treatment is declining. Although putting efforts on the traditional in vivo or in vitro methods, pharmaceutical financial investment has been reduced over the years. Therefore, establishing effective computational methods is decisive to find new leads in a reasonable amount of time. Successful approaches have been presented to solve this problem but seldom protein sequences and structured data are used together. In this paper, we present a deep learning architecture model, which exploits the particular ability of Convolutional Neural Networks (CNNs) to obtain 1D representations from protein sequences (amino acid sequence) and compounds SMILES (Simplified Molecular Input Line Entry System) strings. These representations can be interpreted as features that express local dependencies or patterns that can then be used in a Fully Connected Neural Network (FCNN), acting as a binary classifier. The results achieved demonstrate that using CNNs to obtain representations of the data, instead of the traditional descriptors, lead to improved performance. The proposed end-to-end deep learning method outperformed traditional machine learning approaches in the correct classification of both positive and negative interactions.

Imaging-detected incidental thyroid nodules that undergo surgery: a single-center experience over 1 year.

BACKGROUND AND PURPOSE: Incidental thyroid nodules are commonly seen on imaging, and their work-up can ultimately lead to surgery. We describe characteristics and pathology results of imaging-detected incidental thyroid nodules that underwent surgery. MATERIALS AND METHODS: A retrospective review was performed of 303 patients who underwent thyroid surgery over a 1-year period to identify patients who presented with incidental thyroid nodules on imaging. Medical records were reviewed for the types of imaging studies that led to detection, nodule characteristics, and surgical pathology. RESULTS: Of 303 patients, 208 patients (69%) had surgery for thyroid nodules. Forty-seven of 208 patients (23%) had incidental thyroid nodules detected on imaging. The most common technique leading to detection was CT (47%). All patients underwent biopsy before surgery. The cytology results were nondiagnostic (6%), benign (4%), atypia of undetermined significance or follicular neoplasm of undetermined significance (23%), follicular neoplasm or suspicious for follicular neoplasm (19%), suspicious for malignancy (17%), and diagnostic of malignancy (30%). Surgical pathology was benign in 24 of 47 (51%) cases of incidental thyroid nodules. In the 23 incidental cancers, the most common histologic type was papillary (87%), the mean size was 1.4 cm, and nodal metastases were present in 7 of 23 cases (30%). No incidental cancers on imaging had distant metastases. CONCLUSIONS: Imaging-detected incidental thyroid nodules led to nearly one-fourth of surgeries for thyroid nodules, and almost half were initially detected on CT. Despite indeterminate or suspicious cytology results that lead to surgery, more than half were benign on final pathology. Guidelines for work-up of incidental thyroid nodules detected on CT could help reduce unnecessary investigations and surgery.

Effect of varying contrast material iodine concentration and injection technique on the conspicuity of hepatocellular carcinoma during 64-section MDCT of patients with cirrhosis.

OBJECTIVES: The aim of this study was to compare the intraindividual effects of contrast material with two different iodine concentrations on the conspicuity of hepatocellular carcinoma (HCC) and vascular and hepatic contrast enhancement during multiphasic, 64-section multidetector row CT (MDCT) in patients with cirrhosis using two contrast medium injection techniques. METHODS: Patients were randomly assigned to one of two groups with an equal iodine dose but different contrast material injection techniques: scheme A, fixed injection duration (25 s), and scheme B, fixed injection flow rate (4 ml s(-1)). For each group, patients were randomised to receive both moderate-concentration contrast medium (MCCM) and high-concentration contrast medium (HCCM) during two CT examinations within 3 months. Enhancement of the aorta, liver and portal vein and the tumour-to-liver contrast-to-noise ratio (CNR) were compared between MCCM and HCCM. RESULTS: 30 patients (mean age 59 years; range 45-80 years; 16 patients in scheme A and 14 in scheme B) with a total of 31 confirmed HCC nodules were prospectively enrolled. For scheme B, the mean contrast enhancement of the aorta and tumour-to-liver CNR were significantly higher with HCCM than with MCCM during the hepatic arterial phase (+350.5 HU vs +301.1 HU, p = 0.001, and +7.5 HU vs +5.5 HU, p = 0.004). For both groups, there was no significant difference between MCCM and HCCM for all other comparisons. CONCLUSION: For a constant injection flow rate, HCCM significantly improves the conspicuity of HCC lesions and aortic enhancement during the hepatic arterial phase on 64-section MDCT in patients with cirrhosis.

In vivo visualization of abdominal malignancies with acoustic radiation force elastography.

The utility of acoustic radiation force impulse (ARFI) imaging for real-time visualization of abdominal malignancies was investigated. Nine patients presenting with suspicious masses in the liver (n = 7) or kidney (n = 2) underwent combined sonography/ARFI imaging. Images were acquired of a total of 12 tumors in the nine patients. In all cases, boundary definition in ARFI images was improved or equivalent to boundary definition in B-mode images. Displacement contrast in ARFI images was superior to echo contrast in B-mode images for each tumor. The mean contrast for suspected hepatocellular carcinomas (HCCs) in B-mode images was 2.9 dB (range: 1.5-4.2) versus 7.5 dB (range: 3.1-11.9) in ARFI images, with all HCCs appearing more compliant than regional cirrhotic liver parenchyma. The mean contrast for metastases in B-mode images was 3.1 dB (range: 1.2-5.2) versus 9.3 dB (range: 5.7-13.9) in ARFI images, with all masses appearing less compliant than regional non-cirrhotic liver parenchyma. ARFI image contrast (10.4 dB) was superior to B-mode contrast (0.9 dB) for a renal mass. To our knowledge, we present the first in vivo images of abdominal malignancies in humans acquired with the ARFI method or any other technique of imaging tissue elasticity.

Liver ablation guidance with acoustic radiation force impulse imaging: challenges and opportunities.

Previous studies have established the feasibility of monitoring radiofrequency (RF) ablation procedures with acoustic radiation force impulse (ARFI) imaging. However, questions remained regarding the utility of the technique in clinically realistic scenarios and at scanning depths associated with abdominal imaging in adults. We address several of these issues and detail recent progress towards the clinical relevance of the ARFI technique. Results from in vitro bovine tissues and an in vivo ovine model are presented. Additional experiments were conducted with a tissue-mimicking phantom and parallel receive tracking techniques in order to further support the clinical feasibility of the method. Thermal lesions created during RF ablation are visualized with high contrast in both in vitro and in vivo hepatic tissues, and radial lesion growth can be monitored throughout the duration of the procedure. ARFI imaging is implemented on a diagnostic ultrasonic scanner, and thus may be a convenient option to guide RF ablation procedures, particularly when electrode insertion is also performed with sonographic guidance.

Sonographically guided thrombin injection of iatrogenic femoral pseudoaneurysms: further experience of a single institution.

OBJECTIVE: In September 1998, we began to treat iatrogenic femoral pseudoaneurysms with direct thrombin injection under sonographic guidance. Our purpose was to determine the success and complication rate of this technique. SUBJECTS AND METHODS: We treated 114 consecutive patients who had iatrogenic femoral pseudoaneurysms using direct thrombin injection. A 22-gauge spinal needle was placed into the pseudoaneurysm lumen with sonographic guidance, and bovine or human thrombin (mean dose, 306 U; range, 50--1600 U) was injected under continuous color Doppler sonographic visualization. Distal pulses were monitored. Patient demographics, clinical variables, and pseudoaneurysm characteristics were collected. RESULTS: One hundred three (90%) of 114 patients had pseudoaneurysm thrombosis after the first procedure. Of the remaining 11 patients who required a second procedure 1 day later, thrombosis occurred in seven (64%) of 11. Thus, the overall success rate was 96% (110/114). Of the patients who required one injection, the mean thrombosis time was 12 sec (range, 3--90 sec). Three (3%) of 114 patients required conscious sedation. Of the patients with successful thrombosis, 24-hr follow-up sonograms showed no recurrent pseudoaneurysm. Four patients (4%) had potential complications: a "blue toe" 15 hr after the thrombin injection that resolved spontaneously, a groin abscess, leg ischemia that resolved spontaneously after 4 hr, and crampy buttock pain that resolved spontaneously. CONCLUSION: For the treatment of iatrogenic femoral pseudoaneurysms, thrombin injection under sonographic guidance is a quick and effective method of therapy. Failures and complications are infrequent. At our institution, sonographically guided thrombin injection has replaced compression repair.

Nonstenotic internal carotid arteries: effects of age and blood pressure at the time of scanning on Doppler US velocity measurements.

PURPOSE: To assess the effects of age and blood pressure at the time of scanning on internal carotid artery velocities and cross-sectional diameter at Doppler ultrasonography (US). MATERIALS AND METHODS: During 12 months, 1,020 consecutive patients underwent internal carotid artery Doppler US. No or minimal arterial disease was found in 142 patients (67 women, 75 men). Blood pressure was recorded prior to examination. The angle-corrected internal carotid artery peak systolic and end-diastolic velocities were obtained. The effects of systolic blood pressure, diastolic blood pressure, pulse pressure, age, chronic hypertension, and medications for hypertension on velocities were evaluated by using linear regression analysis. RESULTS: Peak systolic velocity was influenced by age (P =.008), systolic blood pressure (P =.009), diastolic blood pressure (P =.003), and pulse pressure (P =.017) but not history of hypertension (P =.53) or antihypertensive medication use (P =.77). Increasing age decreased peak systolic velocity by 0.34 cm/sec/y. End-diastolic velocity was influenced by age (P <.001) but not by systolic, diastolic, or pulse pressure (all P values were >.13). CONCLUSION: Internal carotid artery peak systolic velocities decrease with advancing age and increase with increasing pulse pressure. The effects of blood pressure at the time of scanning are small, but isolated systolic hypertension could cause increases in spurious velocity.

Compression CT urography: a comparison with IVU in the opacification of the collecting system and ureters.

OBJECTIVE: The purpose of this study was to evaluate opacification of the collecting system and ureters using compression computed tomography (CT) urography compared with conventional intravenous urography (IVU). MATERIALS AND METHODS: Fifty consecutive patients underwent compression CT urography as part of a dedicated renal CT. A compression belt was applied prior to nephrographic phase imaging. Excretory phase scans were acquired through the kidneys 3 minutes post injection with the compression belt in place. The compression belt was then released, and scans were obtained through the ureters. Three independent readers then scored opacification of the collecting system and ureters on a scale of 0-2 (0 = no opacification, 1 = partial opacification, 2 = full opacification and distension). Fifty consecutive nonmatched IVUs were scored by segment by the same readers. Comparison of the two modalities was made using the Mann-Whitney U test. Interobserver agreement was assessed by the Kappa coefficient. RESULTS: CT demonstrated significantly better opacification (p < or = 0.02) of the upper and lower pole pelvicalyceal systems and midureters bilaterally. There was no difference in opacification of the proximal and distal ureters by CT compared with IVU. The Kappa coefficient was 0.53. CONCLUSIONS: Compression CT urography yields equal or better opacification of the collecting system and ureters when compared with IVU, and shows promise for the routine evaluation of the renal excretory system.

Two-dimensional multiplanar and three-dimensional volume-rendered vascular CT in pancreatic carcinoma: interobserver agreement and comparison with standard helical techniques.

OBJECTIVE: The purpose of this study was to compare two-dimensional curved multiplanar and three-dimensional reconstructions, routine axial presentations, and combined techniques in the assessment of vascular involvement by pancreatic malignancy. MATERIALS AND METHODS: For 44 patients with known pancreatic malignancy a total of 56 arterial phase helical CT scans were obtained. Targeted pancreatic imaging was performed, and reformatted images were generated. Axial source images, reformatted images, and the combination of axial and reformatted images were interpreted independently by three observers. The observers graded the celiac axis, common and proper hepatic, splenic, gastroduodenal, and superior mesenteric arteries for tumor involvement. Grades of vascular involvement were compared by intra- and interobserver variability analyses. RESULTS: Intraobserver agreement averaged over five vessels was good between the axial and combined techniques for each individual observer (0.64 < or kappa < or = 0.66), but intraobserver agreement was poor between the axial and reformatted (kappa = 0.17 and kappa = 0.31, respectively) and the reformatted and combined techniques (kappa = 0.31 and kappa = 0.38, respectively) for two observers. For grading of vascular involvement in each vessel, intraobserver agreement was good to excellent between the axial and combined techniques (0.48 or = kappa < or = 0.82). Interobserver agreement averaged over five vessels was poor for imaging techniques except between observer 2 and observer 3 on the axial (kappa = 0.47) and combined techniques (kappa = 0.47). For grading of vascular involvement in each vessel, interobserver agreement for reformatted technique was poor (0.09 < or = kappa < or = 0.40). CONCLUSION: Multiplanar and volume-rendered techniques showed the highest intra- and interobserver variability in grading vascular involvement by pancreatic malignancy. These images should be used in combination with routine axial images to decrease observer variability.

Three-dimensional CT of the genitourinary tract.

The applications of three-dimensional (3D) CT techniques encompass a spectrum from calculus disease to preoperative planning. With proper selection of display windows and levels, accurate measurement of stone size can be achieved, along with volumetric information. A CT scan with reconstruction may help guide the direction of an endopyelotomy incision away from crossing vessels. The benefits of 3D CT in the evaluation of living renal donors include lower cost and decreased patient morbidity. In renal allograft recipients and other patients, the study may be used to investigate hypertension. Also, 3D CT is invaluable in planning nephron-sparing surgery for renal masses. The ultimate role of this modality rests in the hands of clinicians who can benefit from them and the radiologists who must provide the high-quality images and the interpretive expertise.

An electronic device for needle placement during sonographically guided percutaneous intervention.

An electronic device for guiding needle placement during sonographically directed percutaneous intervention was tested in a phantom and then in patients. In the phantom, targeting accuracy was similar for use of the needle guide alone, the needle guide with the device, and freehand techniques with the device, but all were superior to the freehand technique alone (P <.001). In 34 (79%) of 43 patients, the device worked well.

Breath-hold three-dimensional CT of the liver with multi-detector row helical CT.

PURPOSE: To compare image quality on transverse source images and coronal and sagittal reformations to determine the feasibility of using single-breath-hold three-dimensional liver computed tomography (CT) with multi-detector row helical CT in patients suspected of having hepatic metastases. MATERIALS AND METHODS: Fifty-three patients underwent the protocol. Coronal and sagittal reformations were constructed. Images were reviewed for duration of scan acquisition and length and adequacy of z-axis coverage. Reformations were scored for visualization of portal and hepatic vein branches, liver edge sharpness, cardiac pulsation and respiratory motion artifacts, noise due to mottle, and overall impression. RESULTS: Mean z-axis coverage was 207 mm +/- 33 (SD) (range, 145-280 mm), with a mean acquisition time of 10.96 seconds +/- 1.78 (range, 7.73-14.93 seconds). In 44 (83%) patients, the entire liver was imaged on a single helical scan. Artifact from cardiac motion was not identified on the transverse source images in any patient but was identified on coronal images in eight (15%) and on sagittal images in seven (13%). Similarly, noise due to mottle was not identified on the transverse source images but was identified on coronal images in seven (13%) patients and on sagittal images in six (11%). CONCLUSION: It is feasible to perform single-breath-hold three-dimensional liver CT with multi-detector row helical CT technology. Reformations provide a unique perspective with which to view the liver and may improve diagnostic capacity.

Nonenhanced helical CT and US in the emergency evaluation of patients with renal colic: prospective comparison.

PURPOSE: To compare nonenhanced helical computed tomography (CT) with ultrasonography (US) for the depiction of urolithiasis. MATERIALS AND METHODS: During 9 months, 45 patients (mean age, 44 years; mean weight, 92.5 kg) prospectively underwent both nonenhanced helical CT (5-mm collimation; pitch of 1.5) and US of the kidneys, ureters, and bladder. US evaluation included a careful search for ureteral calculi. Presence of calculi and obstruction and incidental diagnoses were recorded. Clinical, surgical, and/or imaging follow-up data were obtained in all patients. The McNemar test was used to compare groups. RESULTS: Diagnoses included 23 ureteral calculi and one each of renal cell carcinoma, appendicitis, ureteropelvic junction obstruction, renal subcapsular hematoma, cholelithiasis, medullary calcinosis, and myelolipoma. CT depicted 22 of 23 ureteral calculi (sensitivity, 96%). US depicted 14 of 23 ureteral calculi (sensitivity, 61%). Differences in sensitivity were statistically significant (P: =.02). Specificity for each technique was 100%. When modalities were compared for the detection of any clinically relevant abnormality (eg, unilateral hydronephrosis and/or urolithiasis in patients with an obstructing calculus), sensitivities of US and CT increased to 92% and 100%, respectively. One case of appendicitis was missed at US, whereas medullary calcinosis and myelolipoma were missed at CT. CONCLUSION: Nonenhanced CT has a higher sensitivity for the detection of ureteral calculi compared with US.

Logistic advantages of four-section helical CT in the abdomen and pelvis.

BACKGROUND: Multisection helical computed tomography (CT) has the potential for providing data sets with better section profiles, more anatomic coverage, and shorter breath-holding periods. Our purpose was to quantitate these advantages in a clinical setting when imaging the abdomen and pelvis. METHODS: CT parameters including collimation, timing, z-axis coverage, and milliamperes were gathered retrospectively for the image set of both single-section (GE CT/i with 0.8-s rotation) and four-section (GE QX/i Lightspeed with 0.8-s rotation) helical CT scanners. Data were recorded for the abdomen and pelvis CT (n = 30 each), dual-phase liver CT including the pelvis (n = 15 each), and dual-phase pancreas CT (n = 15 each). RESULTS: The abdominal and pelvic CT averaged 128.4 +/- 5.4 s for single-section scanners (70-s delay, two breath-holds of 21.1 and 17. 7 s with a 19.5-s interscan delay) and 92.2 +/- 2.2 s for the four-section scanner (70-s delay and a 22.2-s breath-hold; p < 0. 0001). For the dual liver and pelvis CT, single-section scanners averaged 119.9 +/- 7.5 s (30-s delay, 15.8-s arterial phase, 20.0-s interscan delay, 21.2-s venous phase, 19.5-s interscan delay, and 14. 2 s for the remaining abdomen and pelvis), whereas the four-section scanner averaged 86.8 +/- 2.5 s (30-s delay, 6.7-s arterial phase, 27.9-s interscan delay, and 21.8-s venous phase including the pelvis; p < 0.0001). For the dual pancreas CT, single-section scanners averaged 86.7 +/- 2.5 s (20-s delay, 28.3-s arterial phase, 17.8-s interscan delay, 21.7-s venous phase), whereas the four-section scanner averaged 78.0 +/- 2.9 s (20-s delay, 9.7-s arterial phase, 30.7-s interscan delay, 13.0-s venous phase; p < 0. 0001). CONCLUSION: CT scanners having four-section technology can reduce overall data acquisition times by 10-30% and total milliamperes by 50-60% depending on the protocol with thinner slice profiles.