Artificial intelligence in molecular imaging

Herskovits, Edward H.

doi:10.21037/atm-20-6191

Cited by 16 publications

(12 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In MRI (Magnetic Resonance Imaging), for instance, the issue of hallucinations has been discussed in the fastMRI challenge [17], where they are referred to as 'not acceptable' and 'especially problematic'. Similar sentiments have been expressed in the case of microscopy [18], fluorescence microscopy [19], PET (Position Emission Tomography) [20] and computed tomography [21,22]. See also [23][24][25][26][27][28] for further discussion and related issues.…”

Section: Issues With Deep Learning For Inverse Problemsmentioning

confidence: 67%

NESTANets: Stable, accurate and efficient neural networks for analysis-sparse inverse problems

Neyra-Nesterenko¹,

Adcock²

2022

Preprint

View full text Add to dashboard Cite

Solving inverse problems is a fundamental component of science, engineering and mathematics. With the advent of deep learning, deep neural networks have significant potential to outperform existing state-of-theart, model-based methods for solving inverse problems. However, it is known that current data-driven approaches face several key issues, notably instabilities and hallucinations, with potential impact in critical tasks such as medical imaging. This raises the key question of whether or not one can construct stable and accurate deep neural networks for inverse problems. In this work, we present a novel construction of an accurate, stable and efficient neural network for inverse problems with general analysis-sparse models. To construct the network, we unroll NESTA, an accelerated first-order method for convex optimization. Combined with a compressed sensing analysis, we prove accuracy and stability. Finally, a restart scheme is employed to enable exponential decay of the required network depth, yielding a shallower, and consequently more efficient, network. We showcase this approach in the case of Fourier imaging, and verify its stability and performance via a series of numerical experiments. The key impact of this work is to provide theoretical guarantees for computing and developing stable neural networks in practice.

show abstract

Section: Issues With Deep Learning For Inverse Problemsmentioning

confidence: 67%

NESTANets: Stable, accurate and efficient neural networks for analysis-sparse inverse problems

Neyra-Nesterenko¹,

Adcock²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…In this context, we will generally be referring to weak AI, ie, AI algorithms based on neural networks that can learn from existing data and make relevant and accurate determinations when exposed to new data. 150 With AI for automated whole-body image interpretation, disease segmentation, and burden determination, 151 it will be possible to apply the principle of theranostics in powerful ways to improve patient care. Among the many foreseeable applications of the information derived from AI are patient selection for an appropriate theranostic agent, prognostication based on imaging and clinical parameters, and selection of an appropriate dose that balances efficacy with tolerable side effects.…”

Section: Challenges Potential and Future Directionsmentioning

confidence: 99%

“…The increasing use of theranostic agents for managing cancer will dovetail with the use of AI for several relevant applications. In this context, we will generally be referring to weak AI , ie, AI algorithms based on neural networks that can learn from existing data and make relevant and accurate determinations when exposed to new data 150 . With AI for automated whole‐body image interpretation, disease segmentation, and burden determination, 151 it will be possible to apply the principle of theranostics in powerful ways to improve patient care.…”

Section: Challenges Potential and Future Directionsmentioning

confidence: 99%

Molecular imaging in oncology: Current impact and future directions

Rowe

Pomper

2021

CA A Cancer J Clinicians

182

111

View full text Add to dashboard Cite

The authors define molecular imaging, according to the Society of Nuclear Medicine and Molecular Imaging, as the visualization, characterization, and measurement of biological processes at the molecular and cellular levels in humans and other living systems. Although practiced for many years clinically in nuclear medicine, expansion to other imaging modalities began roughly 25 years ago and has accelerated since. That acceleration derives from the continual appearance of new and highly relevant animal models of human disease, increasingly sensitive imaging devices, high-throughput methods to discover and optimize affinity agents to key cellular targets, new ways to manipulate genetic material, and expanded use of cloud computing. Greater interest by scientists in allied fields, such as chemistry, biomedical engineering, and immunology, as well as increased attention by the pharmaceutical industry, have likewise contributed to the boom in activity in recent years. Whereas researchers and clinicians have applied molecular imaging to a variety of physiologic processes and disease states, here, the authors focus on oncology, arguably where it has made its greatest impact. The main purpose of imaging in oncology is early detection to enable interception if not prevention of full-blown disease, such as the appearance of metastases. Because biochemical changes occur before changes in anatomy, molecular imaging-particularly when combined with liquid biopsy for screening purposes-promises especially early localization of disease for optimum management. Here, the authors introduce the ways and indications in which molecular imaging can be undertaken, the tools used and under development, and near-term challenges and opportunities in oncology.

show abstract

“…With this strategy, surgeons can be provided with more precise complementary information of structural imaging and functional imaging, therefore easing surgery development and improving patient prognosis [ 161 ]. Another interesting future research line is related to the application of artificial intelligence (AI) in molecular imaging [ 162 ]. In fact, AI in general and deep learning in particular have, to varying degrees, impacted almost all aspects of molecular imaging, from image acquisition to diagnosis, especially in imaging processing.…”

Section: General Remarks and Future Perspectivesmentioning

confidence: 99%

“…In fact, AI in general and deep learning in particular have, to varying degrees, impacted almost all aspects of molecular imaging, from image acquisition to diagnosis, especially in imaging processing. As a result, great precision in the diagnosis and follow-up of cancer patients has been achieved [ 162 ].…”

Section: General Remarks and Future Perspectivesmentioning

confidence: 99%

Recent Advances in Multimodal Molecular Imaging of Cancer Mediated by Hybrid Magnetic Nanoparticles

et al. 2021

View full text Add to dashboard Cite

Cancer is the second leading cause of death in the world, which is why it is so important to make an early and very precise diagnosis to obtain a good prognosis. Thanks to the combination of several imaging modalities in the form of the multimodal molecular imaging (MI) strategy, a great advance has been made in early diagnosis, in more targeted and personalized therapy, and in the prediction of the results that will be obtained once the anticancer treatment is applied. In this context, magnetic nanoparticles have been positioned as strong candidates for diagnostic agents as they provide very good imaging performance. Furthermore, thanks to their high versatility, when combined with other molecular agents (for example, fluorescent molecules or radioisotopes), they highlight the advantages of several imaging techniques at the same time. These hybrid nanosystems can be also used as multifunctional and/or theranostic systems as they can provide images of the tumor area while they administer drugs and act as therapeutic agents. Therefore, in this review, we selected and identified more than 160 recent articles and reviews and offer a broad overview of the most important concepts that support the synthesis and application of multifunctional magnetic nanoparticles as molecular agents in advanced cancer detection based on the multimodal molecular imaging approach.

show abstract

Artificial intelligence in molecular imaging

Cited by 16 publications

References 69 publications

NESTANets: Stable, accurate and efficient neural networks for analysis-sparse inverse problems

NESTANets: Stable, accurate and efficient neural networks for analysis-sparse inverse problems

Molecular imaging in oncology: Current impact and future directions

Recent Advances in Multimodal Molecular Imaging of Cancer Mediated by Hybrid Magnetic Nanoparticles

Contact Info

Product

Resources

About