Prof. Adam M. Alessio Profile

Prof. Adam M. Alessio

Professor at Michigan State University

SPIE Involvement:

Conference Program Committee | Author

Websites:

Personal Website

Publications (26)

SPIE Journal Paper | 4 June 2024

Near-pair patch generative adversarial network for data augmentation of focal pathology object detection models

Ethan Tu, Jonathan Burkow, Andy Tsai, Joseph Junewick, Francisco A. Perez, Jeffrey Otjen, Adam M. Alessio

JMI, Vol. 11, Issue 03, 034505, (June 2024) https://doi.org/10.1117/12.10.1117/1.JMI.11.3.034505

KEYWORDS: Education and training, Object detection, Radiography, Sensors, Data modeling, Pathology, Gallium nitride, Visualization, Machine learning, Feature extraction

Read Abstract +

PurposeThe limited volume of medical training data remains one of the leading challenges for machine learning for diagnostic applications. Object detectors that identify and localize pathologies require training with a large volume of labeled images, which are often expensive and time-consuming to curate. To reduce this challenge, we present a method to support distant supervision of object detectors through generation of synthetic pathology-present labeled images.ApproachOur method employs the previously proposed cyclic generative adversarial network (cycleGAN) with two key innovations: (1) use of “near-pair” pathology-present regions and pathology-absent regions from similar locations in the same subject for training and (2) the addition of a realism metric (Fréchet inception distance) to the generator loss term. We trained and tested this method with 2800 fracture-present and 2800 fracture-absent image patches from 704 unique pediatric chest radiographs. The trained model was then used to generate synthetic pathology-present images with exact knowledge of location (labels) of the pathology. These synthetic images provided an augmented training set for an object detector.ResultsIn an observer study, four pediatric radiologists used a five-point Likert scale indicating the likelihood of a real fracture (1 = definitely not a fracture and 5 = definitely a fracture) to grade a set of real fracture-absent, real fracture-present, and synthetic fracture-present images. The real fracture-absent images scored 1.7±1.0, real fracture-present images 4.1±1.2, and synthetic fracture-present images 2.5±1.2. An object detector model (YOLOv5) trained on a mix of 500 real and 500 synthetic radiographs performed with a recall of 0.57±0.05 and an F2 score of 0.59±0.05. In comparison, when trained on only 500 real radiographs, the recall and F2 score were 0.49±0.06 and 0.53±0.06, respectively.ConclusionsOur proposed method generates visually realistic pathology and that provided improved object detector performance for the task of rib fracture detection.

Proceedings Article | 3 April 2024 Poster + Paper

Deep learning methods for multi-class pneumoconioses grading of chest radiographs

Meiqi Liu, Ian Loveless, Zenas Huang, Kenneth Rosenman, Ling Wang, Adam Alessio

Proceedings Volume 12927, 129272N (2024) https://doi.org/10.1117/12.3006266

KEYWORDS: Computer aided detection, Chest imaging, Radiomics, Radiography, Deep learning, Feature extraction, Data modeling, Lung

Read Abstract +

Proceedings Article | 4 April 2022 Poster + Paper

Avalanche decision schemes to improve pediatric rib fracture detection

Jonathan Burkow, Gregory Holste, Jeffrey Otjen, Francisco Perez, Joseph Junewick, Adam Alessio

Proceedings Volume 12033, 120332A (2022) https://doi.org/10.1117/12.2611013

KEYWORDS: Performance modeling, Chest imaging, Radiography, X-ray computed tomography, Visualization, X-rays, Convolutional neural networks, Object recognition

Read Abstract +

Proceedings Article | 15 February 2021 Presentation + Paper

Machine learned approach for estimating myocardial blood flow from dynamic CT and coronary artery disease risk factors

Ethan Tu, Muneeza Azmat, Kelley Branch, Adam Alessio

Proceedings Volume 11600, 116001I (2021) https://doi.org/10.1117/12.2581703

KEYWORDS: Solid modeling, Positron emission tomography, Blood circulation, Arteries, Machine learning, Data modeling, Principal component analysis, Ischemia, Factor analysis, Signal attenuation

Read Abstract +

Proceedings Article | 15 February 2021 Presentation + Paper

Machine learned versus analytical models for estimation of fractional flow reserve from CT-derived information

Muneeza Azmat, Ethan Tu, Kelley Branch, Adam Alessio

Proceedings Volume 11600, 116000W (2021) https://doi.org/10.1117/12.2581701

KEYWORDS: Data modeling, Statistical analysis, Machine learning, Blood circulation, Process modeling, General packet radio service, Error analysis, Arteries, Transducers, Performance modeling

Read Abstract +

Fractional Flow Reserve (FFR) is a widely used metric to quantify the functional significance of stenoses in coronary arteries. FFR is the ratio of pressure before and after a stenosis and is measured using a transducer during coronary catheterization. To avoid unnecessary catheterization, many analytical and data-driven models based on non-invasive imaging have been proposed for FFR estimation. In this study, we construct physics-informed analytical models and datadriven machine learning models for FFR estimation based on CT-derived information. All four models require simple information about suspect stenoses, offering rapid, practical approaches for functional assessment. The four models we study are: (1) a patient-specific blood flow informed pressure drop model based on Navier-Stokes equations, (2) a purely geometric model based on stenosis area reduction, (3) a Gaussian process regression model trained on patient specific stenosis geometry and blood flow data, and (4) a Gaussian process regression model trained only on patient specific stenosis geometry. The models were developed and tested using a simulation study with ground truth FFR values from computational fluid dynamics analysis of blood flow through a population of stenosed arteries. In total, 60 different stenosis conditions based on known prevalence were simulated with a range of measurement errors leading to 10000 data sets. The RMSE of the model estimates for approach 1 through 4 are, 0.19, 0.45, 0.06, 0.14. The flow informed machine learning model leads to ~1% lower bias and ~12% lower variance than the flow informed analytical model. Considering the improved variance performance, the machine learning models likely outperform analytic expressions because they learn optimal regression associations robust to noise. This work suggests that machine learned approaches may be superior to conventional analytic expressions for FFR estimation, particularly when inputs contain realistic measurement error.

Proceedings Article | 16 March 2020 Paper

Genetic algorithm for machine learning architecture selection for breast MRI classification

Peter Carras, Carina Pereira, Debosmita Biswas, Christoph Lee, Savannah Partridge, Adam Alessio

Proceedings Volume 11314, 1131422 (2020) https://doi.org/10.1117/12.2547490

KEYWORDS: Genetic algorithms, Breast, Magnetic resonance imaging, Data modeling, Machine learning, Image classification, Performance modeling, Convolutional neural networks, Network architectures

Read Abstract +

Proceedings Article | 16 March 2020 Paper

Hyperparameter selection for ResNet classification of malignancy from thyroid ultrasound images

Joseph Cox, Sydney Rubin, Joe Adams, Carina Pereira, Manjiri Dighe, Adam Alessio

Proceedings Volume 11314, 1131447 (2020) https://doi.org/10.1117/12.2550531

KEYWORDS: Data modeling, RGB color model, Ultrasonography, Performance modeling, Image classification, Elastography, Biopsy, Fine needle aspiration, Optimization (mathematics)

Read Abstract +

Proceedings Article | 16 March 2020 Paper

Deep learning methods for segmentation of lines in pediatric chest radiographs

Ryan Sullivan, Greg Holste, Jonathan Burkow, Adam Alessio

Proceedings Volume 11314, 113142I (2020) https://doi.org/10.1117/12.2550686

KEYWORDS: Image segmentation, Computer programming, Chest imaging, Statistical modeling, Data modeling, Performance modeling, Image classification

Read Abstract +

Proceedings Article | 10 March 2020 Presentation + Paper

Multi-class semantic segmentation of pediatric chest radiographs

Gregory Holste, Ryan Sullivan, Michael Bindschadler, Nicholas Nagy, Adam Alessio

Proceedings Volume 11313, 113131E (2020) https://doi.org/10.1117/12.2544426

KEYWORDS: Image segmentation, Radiography, Chest imaging, Performance modeling, Data modeling, Chest, Spine, Lung, Binary data

Read Abstract +

Proceedings Article | 9 March 2018 Presentation + Paper

Evaluation of radiation dose reduction via myocardial frame reduction in dynamic cardiac CT for perfusion quantitation

Michael Bindschadler, Kelley Branch, Adam Alessio

Proceedings Volume 10573, 105731W (2018) https://doi.org/10.1117/12.2294556

KEYWORDS: Heart, Data acquisition, Error analysis, Computed tomography, Data modeling, X-ray computed tomography, Statistical analysis, Aorta, Cardiac imaging

Read Abstract +

Proceedings Article | 27 February 2018 Presentation + Paper

Comparison of machine learned approaches for thyroid nodule characterization from shear wave elastography images

Carina Pereira, Manjiri Dighe, Adam Alessio

Proceedings Volume 10575, 105751X (2018) https://doi.org/10.1117/12.2294572

KEYWORDS: Ultrasonography, Machine learning, Feature extraction, Elastography, Tissues, Image classification, Computer aided diagnosis and therapy, Convolutional neural networks, CAD systems, Medical imaging

Read Abstract +

SPIE Journal Paper | 13 May 2017

Variable temporal sampling and tube current modulation for myocardial blood flow estimation from dose-reduced dynamic computed tomography

Dimple Modgil, Michael Bindschadler, Adam Alessio, Patrick La Rivière

JMI, Vol. 4, Issue 02, 026002, (May 2017) https://doi.org/10.1117/12.10.1117/1.JMI.4.2.026002

KEYWORDS: Computed tomography, Blood circulation, Modulation, Error analysis, Arteries

Read Abstract +

Proceedings Article | 10 March 2017 Paper

A phantom design and assessment of lesion detectability in PET imaging

Scott Wollenweber, Paul Kinahan, Adam Alessio

Proceedings Volume 10136, 101361E (2017) https://doi.org/10.1117/12.2254430

KEYWORDS: Positron emission tomography, Imaging systems, Systems modeling, Scanners, Signal to noise ratio, Signal detection, Image quality, Ranging, Data modeling, Data acquisition

Read Abstract +

Proceedings Article | 9 March 2017 Presentation + Paper

Statistical distributions of ultra-low dose CT sinograms and their fundamental limits

Tzu-Cheng Lee, Ruoqiao Zhang, Adam Alessio, Lin Fu, Bruno De Man, Paul Kinahan

Proceedings Volume 10132, 101320N (2017) https://doi.org/10.1117/12.2254375

KEYWORDS: Interference (communication), Aluminum, Computer simulations, Image registration, Sensors, Statistical analysis, Signal attenuation, Computed tomography, Image restoration, Data processing, Calibration, Data modeling, Environmental sensing, Diagnostics, Statistical modeling

Read Abstract +

Proceedings Article | 9 March 2017 Presentation + Paper

Improved attenuation correction for respiratory gated PET/CT with extended-duration cine CT: a simulation study

Ruoqiao Zhang, Adam Alessio, Larry Pierce, Darrin Byrd, Tzu-Cheng Lee, Bruno De Man, Paul Kinahan

Proceedings Volume 10132, 101321I (2017) https://doi.org/10.1117/12.2254599

KEYWORDS: Computed tomography, Positron emission tomography, Medical imaging, Algorithm development, Computer simulations, Lung, Tumors, 3D acquisition, CT reconstruction, Reconstruction algorithms

Read Abstract +

SPIE Journal Paper | 2 May 2016

Evaluation of static and dynamic perfusion cardiac computed tomography for quantitation and classification tasks

Michael Bindschadler, Dimple Modgil, Kelley Branch, Patrick La Rivière, Adam Alessio

JMI, Vol. 3, Issue 02, 024001, (May 2016) https://doi.org/10.1117/12.10.1117/1.JMI.3.2.024001

KEYWORDS: Computed tomography, Computer simulations, Error analysis, Ischemia, Image segmentation, Blood circulation, Angiography, Single photon emission computed tomography, Radiation effects, Optical simulations

Read Abstract +

Proceedings Article | 20 March 2015 Paper

Adaptive sampling of CT data for myocardial blood flow estimation from dose-reduced dynamic CT

Dimple Modgil, Michael Bindschadler, Adam Alessio, Patrick La Rivière

Proceedings Volume 9413, 941304 (2015) https://doi.org/10.1117/12.2082258

KEYWORDS: Blood circulation, Computed tomography, Iodine, Data modeling, Arteries, Computer aided diagnosis and therapy, Error analysis, Radiology, Signal attenuation, Image analysis

Read Abstract +

Proceedings Article | 18 March 2015 Paper

Performance comparison between static and dynamic cardiac CT on perfusion quantitation and patient classification tasks

Michael Bindschadler, Dimple Modgil, Kelley Branch, Patrick La Riviere, Adam Alessio

Proceedings Volume 9412, 941224 (2015) https://doi.org/10.1117/12.2082098

KEYWORDS: Computed tomography, Computer simulations, Data modeling, Visualization, Blood circulation, Ischemia, X-ray computed tomography, Physics, Image segmentation, Heart

Read Abstract +

SPIE Journal Paper | 3 November 2014

Sinogram smoothing techniques for myocardial blood flow estimation from dose-reduced dynamic computed tomography

Dimple Modgil, Adam Alessio, Michael Bindschadler, Patrick La Rivière

JMI, Vol. 1, Issue 03, 034004, (November 2014) https://doi.org/10.1117/12.10.1117/1.JMI.1.3.034004

KEYWORDS: Computed tomography, Blood circulation, Iodine, Smoothing, Data modeling, Error analysis, Sensors, Positron emission tomography, Single photon emission computed tomography, Image restoration

Read Abstract +

SPIE Journal Paper | 24 September 2014

Assessment of patient selection criteria for quantitative imaging with respiratory-gated positron emission tomography

Stephen Bowen, Larry Pierce, Adam Alessio, Chi Liu, Scott Wollenweber, Charles Stearns, Paul Kinahan

JMI, Vol. 1, Issue 02, 026001, (September 2014) https://doi.org/10.1117/12.10.1117/1.JMI.1.2.026001

KEYWORDS: Positron emission tomography, Lung, Liver, Statistical analysis, Motion models, Computed tomography, Diagnostics, Signal attenuation, Medicine, Image acquisition

Read Abstract +

Proceedings Article | 21 March 2014 Paper

Adaptive temporal smoothing of sinogram data using Karhunen-Loeve (KL) transform for myocardial blood flow estimation from dose-reduced dynamic CT

Dimple Modgil, Adam Alessio, Michael Bindschadler, Patrick La Rivière

Proceedings Volume 9034, 903422 (2014) https://doi.org/10.1117/12.2044185

KEYWORDS: Blood circulation, Smoothing, Iodine, Computed tomography, Signal attenuation, Data modeling, Arteries, Computer aided diagnosis and therapy, Image filtering, Bone

Read Abstract +

Proceedings Article | 19 March 2014 Paper

Simulation evaluation of quantitative myocardial perfusion assessment from cardiac CT

Michael Bindschadler, Dimple Modgil, Kelley Branch, Patrick La Riviere, Adam Alessio

Proceedings Volume 9033, 903303 (2014) https://doi.org/10.1117/12.2043563

KEYWORDS: Iodine, Blood circulation, Signal attenuation, Error analysis, Tissues, Capillaries, Data acquisition, X-rays, Computed tomography, Optical simulations

Read Abstract +

Contrast enhancement on cardiac CT provides valuable information about myocardial perfusion and methods have been proposed to assess perfusion with static and dynamic acquisitions. There is a lack of knowledge and consensus on the appropriate approach to ensure 1) sufficient diagnostic accuracy for clinical decisions and 2) low radiation doses for patient safety. This work developed a thorough dynamic CT simulation and several accepted blood flow estimation techniques to evaluate the performance of perfusion assessment across a range of acquisition and estimation scenarios. Cardiac CT acquisitions were simulated for a range of flow states (Flow = 0.5, 1, 2, 3 ml/g/min, cardiac output = 3,5,8 L/min). CT acquisitions were simulated with a validated CT simulator incorporating polyenergetic data acquisition and realistic x-ray flux levels for dynamic acquisitions with a range of scenarios including 1, 2, 3 sec sampling for 30 sec with 25, 70, 140 mAs. Images were generated using conventional image reconstruction with additional image-based beam hardening correction to account for iodine content. Time attenuation curves were extracted for multiple regions around the myocardium and used to estimate flow. In total, 2,700 independent realizations of dynamic sequences were generated and multiple MBF estimation methods were applied to each of these. Evaluation of quantitative kinetic modeling yielded blood flow estimates with an root mean square error (RMSE) of ~0.6 ml/g/min averaged across multiple scenarios. Semi-quantitative modeling and qualitative static imaging resulted in significantly more error (RMSE = ~1.2 and ~1.2 ml/min/g respectively). For quantitative methods, dose reduction through reduced temporal sampling or reduced tube current had comparable impact on the MBF estimate fidelity. On average, half dose acquisitions increased the RMSE of estimates by only 18% suggesting that substantial dose reductions can be employed in the context of quantitative myocardial blood flow estimation. In conclusion, quantitative model-based dynamic cardiac CT perfusion assessment is capable of accurately estimating MBF across a range of cardiac outputs and tissue perfusion states, outperforms comparable static perfusion estimates, and is relatively robust to noise and temporal subsampling.

Proceedings Article | 9 March 2011 Paper

Comparison of models and acquisition techniques for estimation of myocardial blood flow from CT

Adam Alessio, Kelley Branch, James Caldwell, James Bassingthwaighte

Proceedings Volume 7965, 79651I (2011) https://doi.org/10.1117/12.878209

KEYWORDS: Blood circulation, Computed tomography, Tissues, Iodine, Picosecond phenomena, Heart, Motion models, Capillaries, Data analysis, Cardiac imaging

Read Abstract +

Proceedings Article | 8 February 2011 Paper

Constrain static target kinetic iterative image reconstruction for 4D cardiac CT imaging

Adam Alessio, Patrick La Riviere

Proceedings Volume 7873, 78730S (2011) https://doi.org/10.1117/12.884662

KEYWORDS: Computed tomography, Image restoration, Signal to noise ratio, Reconstruction algorithms, Heart, Data modeling, Composites, Data acquisition, CT reconstruction, Expectation maximization algorithms

Read Abstract +

Proceedings Article | 22 March 2010 Paper

Noise and bias properties of monoenergetic images from DECT used for attenuation correction with PET/CT and SPECT/CT

Ting Xia, Adam Alessio, Paul Kinahan

Proceedings Volume 7622, 762225 (2010) https://doi.org/10.1117/12.844075

KEYWORDS: Signal attenuation, Single photon emission computed tomography, Positron emission tomography, Computed tomography, Scanners, Aluminum, Calibration, X-ray computed tomography, X-rays, Image processing

Read Abstract +

Proceedings Article | 29 April 2005 Paper

Improved quantitation for PET/CT image reconstruction with system modeling and anatomical priors

Adam Alessio, Paul Kinahan, Thomas Lewellen

Proceedings Volume 5747, (2005) https://doi.org/10.1117/12.595372

KEYWORDS: Systems modeling, Tumors, Positron emission tomography, Reconstruction algorithms, Sensors, Scanners, Data modeling, Image segmentation, Image restoration, Imaging systems

Read Abstract +