Abstract Background Virtual imaging trials (VIT) have made significant advancements through the development of realistic human anatomy models, scanner‐specific simulations, and virtual image interpretation. To promote VIT widespread adoption in medical community, it is important to develop methods that unify facilitate use VITs, ensuring their reliable application across various studies. Purpose We developed a containerized environment enhance collaboration interoperability platforms. This...

Virtual clinical trials (VCTs) of medical imaging require realistic models human anatomy. For VCTs in breast imaging, a multi-scale Perlin noise method is proposed to simulate anatomical structures tissue the context an ongoing phantom development effort. Four distributions were used replace voxels representing compartments and Cooper's ligaments phantoms. Digital mammography tomosynthesis projections simulated using DBT system configuration. Power-spectrum analyses higher-order statistics...

Abstract A key challenge for the development and deployment of artificial intelligence (AI) solutions in radiology is solving associated data limitations. Obtaining sufficient representative patient datasets with appropriate annotations may be burdensome due to high acquisition cost, safety limitations, privacy restrictions, or low disease prevalence rates. In silico offers a number potential advantages data, such as diminished harm, reduced simplified acquisition, scalability, improved...

<bold>Purpose</bold>: With three-dimensional (3-D) images displayed as stacks of 2-D images, radiologists rely more heavily on vision away from their fixation point to visually process information, guide eye movements, and detect abnormalities. Thus the ability targets point, commonly characterized useful field view (UFOV), becomes critical for these 3-D imaging modalities. We investigate how UFOV, defined eccentricity, in which detection performance degrades a given probability, varies...

The field of medical image quality has relied on the assumption that metrics for simple visual detection tasks are a reliable proxy more clinically realistic search tasks. Rank order signal detectability across conditions often generalizes from to Here, we argue in 3D images represents paradigm shift imaging: radiologists typically cannot exhaustively scrutinize all regions interest with high acuity fovea requiring signals extra-foveal areas (visual periphery) human retina. We hypothesize...

Abstract Objective. In this article, we introduce a computational model for simulating the growth of breast cancer lesions accounting stiffness surrounding anatomical structures. Approach. our model, ligaments are classified as most rigid structures while softer parts occupied by fat and glandular tissues As result these variations in tissue elasticity, rapidly proliferating tumor cells met with differential resistance. It is found that likely to circumvent stiffer terrains such ligaments,...

Three dimensional image modalities introduce a new paradigm for visual search requiring exploration of larger space than 2D imaging modalities. The large number slices in the 3D volumes and limited reading times make it difficult radiologists to explore thoroughly by fixating with their high resolution fovea on all regions each slice. Thus, images, observers must rely much more periphery (points away from fixation) process information. We previously found dissociation signal detectability...

We evaluate 3D search requires model observers that take into account the peripheral human visual processing (foveated models) to predict observer performance. show two different tasks, free and location-known detection, influence relative detectability of signals sizes in synthetic backgrounds mimicking noise found digital breast tomosynthesis. One resembled a microcalcification (a small bright sphere), while other one was designed look like mass larger Gaussian blob). evaluated current...

To generate evidence regarding the safety and efficacy of artificial intelligence (AI) enabled medical devices, AI models need to be evaluated on a diverse population patient cases, some which may not readily available. We propose an evaluation approach for testing imaging that relies in silico pipelines stochastic digital human anatomy (in object space) with without pathology are imaged using replica acquisition system realistic synthetic image datasets. Here, we release M-SYNTH, dataset...

In this study, six biomechanical models for simulating lamb liver behaviour are presented. They validated using similarity coefficients from Medical Image on reconstructed volumes computerised tomography images. particular, the Jaccard and Hausdorff used. Loads of 20 40 g applied to livers their deformation is simulated by means finite element method. The used a linear elastic model, neo-Hookean Mooney-Rivlin an Ogden viscoelastic model viscohyperelastic model. that provided closest reality...

The surgical interventions of the breast are a delicate process where surgeon must be very accurate. patient is in supine position during intervention, however only resource that can used Magnetic Resonance Image (MRI) taken different time and prone position. work presented this paper develops tool transforms MRI into new virtual corresponding to operating table. table captured with Time-of-Flight camera mechanical behavior simulated using biomechanical model takes account tissues form...

Virtual Clinical Trials (VCTs) of breast imaging have been used as a tool for the evaluation and optimization novel systems through computer simulations anatomy, image acquisition, interpretation. VCTs offer significant advantages over clinical trials in terms cost, duration, radiation risk. The performance depends on selection simulated breasts to represent population interest. We developed method selecting populations software phantoms match distribution compressed thickness percent...

Augmented reality (AR) blends the digital and physical worlds by overlapping a virtual image onto see-through environment. However, contrast reduction noise superposition in an AR head-mounted display (HMD) can substantially limit quality human perceptual performance both spaces. To assess AR, we performed model observer studies for various imaging tasks with targets placed worlds. A target detection was developed complete system including optical see-through. Target using different models...

Medical imaging is quickly evolving towards 3D image modalities such as computed tomography (CT), magnetic resonance (MRI) and digital breast tomosynthesis (DBT). These add volumetric information but further increase the need for radiologists to search through data set. Although much known about strategies in 2D images less functional consequences of different strategies. We instructed readers use two strategies: drillers had their eye movements restricted a few regions while they scrolled...

Model observers have a long history of success in predicting human observer performance clinically-relevant detection tasks. New 3D image modalities provide more signal information but vastly increase the search space to be scrutinized. Here, we compared standard linear model (ideal observers, non-pre-whitening matched filter with eye filter, and various versions Channelized Hotelling models) searching 1/f$^{2.8}$ filtered noise images assessed its relationship traditional location known...

Current medical imaging increasingly relies on 3D volumetric data making it difficult for radiologists to thoroughly search all regions of the volume. In some applications (e.g., Digital Breast Tomosynthesis), is typically paired with a synthesized 2D image (2D-S) generated from corresponding We investigate how this pairing affects spatially large and small signals. Observers searched these signals in volumes, 2D-S images, while viewing both. hypothesize that lower spatial acuity observers'...