Artificial intelligence in tumor subregion analysis based on medical imaging: A review

Lin, Mingquan; Wynne, Jacob; Zhou, Boran; Wang, Tonghe; Lei, Yang; Liu, Tian; Yang, Xiaofeng

doi:10.1002/acm2.13321

Cited by 23 publications

(14 citation statements)

References 176 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to ve supervised machine learning algorithms, we subjected our ow cytometry-derived dataset (the standardized training dataset) to two unsupervised clustering algorithms (k-means clustering and Gaussian mixture clustering) in parallel 33 . As shown in Supplementary Fig.…”

Section: Discussionmentioning

confidence: 99%

Machine Learning-based Approaches for Identifying Human Cells Harboring Fetal Chromatin Domain Ablations

Zaheri

Nguyen

et al. 2021

Preprint

View full text Add to dashboard Cite

Two common hemoglobinopathies, sickle cell disease (SCD) and β-thalassemia, arise from genetic mutations within the β-globin gene. A 500-bp motif termed Fetal Chromatin Domain (FCD), upstream of human ϒ-globin locus, may function as a transcriptional regulatory element driving inhibition of the ϒ-globin gene. Here, we hypothesize that the removal of this motif using CRISPR technology may reactivate the expression of ϒ-globin and subsequently restore fetal hemoglobin functionality. In this work we present two different cell morphology-based machine learning approaches that can be used identify cells that harbor FCD genetic modifications. Three candidate models from the first, which uses multilayer perceptron algorithm (MLP 20–26, MLP26-18, and MLP 30 − 26) and flow cytometry-derived cellular data, yielded 0.83 precision, 0.80 recall, 0.82 accuracy, and 0.90 area under the ROC (receiver operating characteristic) curve when predicting the edited cells. In comparison, the candidate model from the second approach, which uses deep learning (T2D5) and DIC microscopy-derived imaging data, performed with less accuracy (0.80) and ROC AUC (0.87). We envision both assays could be valuable and complementary to currently available genotyping protocols for specific genetic modifications which result in morphological changes in human cells.

show abstract

Section: Discussionmentioning

confidence: 99%

Machine Learning-based Approaches for Identifying Human Cells Harboring Fetal Chromatin Domain Ablations

Zaheri

Nguyen

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Neuropilins are the transmembrane glycoproteins mainly located in the growing neurons which help in neuron development. Neuropilin 1 (NRP1) and neuropilin 2 (NRP2) are the receptors of class III semaphorin receptors and are also targeted by several ligands responsible for angiogenesis (Lin et al 2021). Neuropathological analysis with Covid-19 autopsies has revealed that cleaved S protein interacts with NRP1 and NRP2, which in-turn help the virus to infect the nervous system of the human being (Daly et al 2020;Cantuti-Castelvetri et al 2020).…”

Section: The Basic Mechanism Of Entrymentioning

confidence: 99%

Natural therapeutics against SARS CoV2: the potentiality and challenges

Bhar

Jain

Das

2022

Vegetos

View full text Add to dashboard Cite

The incidence of the COVID-19 pandemic completely reoriented global socio-economic parameters and human civilization have experienced the worst situation in the recent past. The rapid mutation rates in viruses have continuously been creating emerging variants of concerns (VOCs) which devastated different parts of the world with subsequent waves of infection. Although, series of antiviral drugs and vaccines were formulated but cent percent effectiveness of these drugs is still awaited. Many of these drugs have different side effects which necessitate proper trial before release. Plants are the storehouse of antimicrobial metabolites which have also long been utilized as traditional medicines against different viral infections. Although, proper mechanism of action of these traditional medicines are unknown, they may be a potential source of effective anti-COVID drug for future implications. Advanced bioinformatic applications have opened up a new arena in predicting these repurposed drugs as a potential COVID mitigator. The present review summarizes brief accounts of the corona virus with their possible entry mechanism. This study also tries to classify different possible anti COVID-19 plant-derived metabolites based on their probable mode of action. This review will surely provide useful information on repurposed drugs to combat COVID-19 in this critical situation.

show abstract

“…To remedy this, convolutional neural networks (CNNs) have been employed to automate the segmentation task. [8][9][10][11] In this study, we propose a novel deep learning-based method to automatically segment the prostate and DILs on MRI for DIL dose boosting radiation therapy. This article is a continuation of our previous conference paper.…”

Section: Introductionmentioning

confidence: 99%

“…Therefore, simple intensity‐based image segmentation methods would be inefficient in distinguishing the specific DILs in MRI. To remedy this, convolutional neural networks (CNNs) have been employed to automate the segmentation task 8–11 …”

Section: Introductionmentioning

confidence: 99%

MRI‐based prostate and dominant lesion segmentation using cascaded scoring convolutional neural network

et al. 2022

Self Cite

View full text Add to dashboard Cite

Purpose Dose escalation to dominant intraprostatic lesions (DILs) is a novel treatment strategy to improve the treatment outcome of prostate radiation therapy. Treatment planning requires accurate and fast delineation of the prostate and DILs. In this study, a 3D cascaded scoring convolutional neural network is proposed to automatically segment the prostate and DILs from MRI. Methods and materials The proposed cascaded scoring convolutional neural network performs end‐to‐end segmentation by locating a region‐of‐interest (ROI), identifying the object within the ROI, and defining the target. A scoring strategy, which is learned to judge the segmentation quality of DIL, is integrated into cascaded convolutional neural network to solve the challenge of segmenting the irregular shapes of the DIL. To evaluate the proposed method, 77 patients who underwent MRI and PET/CT were retrospectively investigated. The prostate and DIL ground truth contours were delineated by experienced radiologists. The proposed method was evaluated with fivefold cross‐validation and holdout testing. Results The average centroid distance, volume difference, and Dice similarity coefficient (DSC) value for prostate/DIL are 4.3 ± 7.5/3.73 ± 3.78 mm, 4.5 ± 7.9/0.41 ± 0.59 cc, and 89.6 ± 8.9/84.3 ± 11.9%, respectively. Comparable results were obtained in the holdout test. Similar or superior segmentation outcomes were seen when compared the results of the proposed method to those of competing segmentation approaches. Conclusions The proposed automatic segmentation method can accurately and simultaneously segment both the prostate and DILs. The intended future use for this algorithm is focal boost prostate radiation therapy.

show abstract

Artificial intelligence in tumor subregion analysis based on medical imaging: A review

Cited by 23 publications

References 176 publications

Machine Learning-based Approaches for Identifying Human Cells Harboring Fetal Chromatin Domain Ablations

Machine Learning-based Approaches for Identifying Human Cells Harboring Fetal Chromatin Domain Ablations

Natural therapeutics against SARS CoV2: the potentiality and challenges

MRI‐based prostate and dominant lesion segmentation using cascaded scoring convolutional neural network

Contact Info

Product

Resources

About