AI-Enhanced Detection of Clinically Relevant Structural and Functional Anomalies in MRI: Traversing the Landscape of Conventional to Explainable Approaches.

Anomaly detection in medical imaging, particularly within the realm of magnetic resonance imaging (MRI), stands as a vital area of research with far-reaching implications across various medical fields. This review meticulously examines the integration of artificial intelligence (AI) in anomaly detection for MR images, spotlighting its transformative impact on medical diagnostics. We delve into the forefront of AI applications in MRI, exploring advanced machine learning (ML) and deep learning (DL) methodologies that are pivotal in enhancing the precision of diagnostic processes. The review provides a detailed analysis of preprocessing, feature extraction, classification, and segmentation techniques, alongside a comprehensive evaluation of commonly used metrics. Further, this paper explores the latest developments in ensemble methods and explainable AI, offering insights into future directions and potential breakthroughs. This review synthesizes current insights, offering a valuable guide for researchers, clinicians, and medical imaging experts. It highlights AI's crucial role in improving the precision and speed of detecting key structural and functional irregularities in MRI. Our exploration of innovative techniques and trends furthers MRI technology development, aiming to refine diagnostics, tailor treatments, and elevate patient care outcomes. LEVEL OF EVIDENCE: 5 TECHNICAL EFFICACY: Stage 1.

A Deep Learning Approach to Diagnostic Classification of Prostate Cancer Using Pathology-Radiology Fusion.

BACKGROUND: A definitive diagnosis of prostate cancer requires a biopsy to obtain tissue for pathologic analysis, but this is an invasive procedure and is associated with complications. PURPOSE: To develop an artificial intelligence (AI)-based model (named AI-biopsy) for the early diagnosis of prostate cancer using magnetic resonance (MR) images labeled with histopathology information. STUDY TYPE: Retrospective. POPULATION: Magnetic resonance imaging (MRI) data sets from 400 patients with suspected prostate cancer and with histological data (228 acquired in-house and 172 from external publicly available databases). FIELD STRENGTH/SEQUENCE: 1.5 to 3.0 Tesla, T2-weighted image pulse sequences. ASSESSMENT: MR images reviewed and selected by two radiologists (with 6 and 17 years of experience). The patient images were labeled with prostate biopsy including Gleason Score (6 to 10) or Grade Group (1 to 5) and reviewed by one pathologist (with 15 years of experience). Deep learning models were developed to distinguish 1) benign from cancerous tumor and 2) high-risk tumor from low-risk tumor. STATISTICAL TESTS: To evaluate our models, we calculated negative predictive value, positive predictive value, specificity, sensitivity, and accuracy. We also calculated areas under the receiver operating characteristic (ROC) curves (AUCs) and Cohen's kappa. RESULTS: Our computational method (https://github.com/ih-lab/AI-biopsy) achieved AUCs of 0.89 (95% confidence interval [CI]: [0.86-0.92]) and 0.78 (95% CI: [0.74-0.82]) to classify cancer vs. benign and high- vs. low-risk of prostate disease, respectively. DATA CONCLUSION: AI-biopsy provided a data-driven and reproducible way to assess cancer risk from MR images and a personalized strategy to potentially reduce the number of unnecessary biopsies. AI-biopsy highlighted the regions of MR images that contained the predictive features the algorithm used for diagnosis using the class activation map method. It is a fully automatic method with a drag-and-drop web interface (https://ai-biopsy.eipm-research.org) that allows radiologists to review AI-assessed MR images in real time. LEVEL OF EVIDENCE: 1 TECHNICAL EFFICACY STAGE: 2.

Artificial intelligence in neuroradiology: a scoping review of some ethical challenges.

Artificial intelligence (AI) has great potential to increase accuracy and efficiency in many aspects of neuroradiology. It provides substantial opportunities for insights into brain pathophysiology, developing models to determine treatment decisions, and improving current prognostication as well as diagnostic algorithms. Concurrently, the autonomous use of AI models introduces ethical challenges regarding the scope of informed consent, risks associated with data privacy and protection, potential database biases, as well as responsibility and liability that might potentially arise. In this manuscript, we will first provide a brief overview of AI methods used in neuroradiology and segue into key methodological and ethical challenges. Specifically, we discuss the ethical principles affected by AI approaches to human neuroscience and provisions that might be imposed in this domain to ensure that the benefits of AI frameworks remain in alignment with ethics in research and healthcare in the future.

A non-invasive artificial intelligence approach for the prediction of human blastocyst ploidy: a retrospective model development and validation study.

BACKGROUND: One challenge in the field of in-vitro fertilisation is the selection of the most viable embryos for transfer. Morphological quality assessment and morphokinetic analysis both have the disadvantage of intra-observer and inter-observer variability. A third method, preimplantation genetic testing for aneuploidy (PGT-A), has limitations too, including its invasiveness and cost. We hypothesised that differences in aneuploid and euploid embryos that allow for model-based classification are reflected in morphology, morphokinetics, and associated clinical information. METHODS: In this retrospective study, we used machine-learning and deep-learning approaches to develop STORK-A, a non-invasive and automated method of embryo evaluation that uses artificial intelligence to predict embryo ploidy status. Our method used a dataset of 10 378 embryos that consisted of static images captured at 110 h after intracytoplasmic sperm injection, morphokinetic parameters, blastocyst morphological assessments, maternal age, and ploidy status. Independent and external datasets, Weill Cornell Medicine EmbryoScope+ (WCM-ES+; Weill Cornell Medicine Center of Reproductive Medicine, NY, USA) and IVI Valencia (IVI Valencia, Health Research Institute la Fe, Valencia, Spain) were used to test the generalisability of STORK-A and were compared measuring accuracy and area under the receiver operating characteristic curve (AUC). FINDINGS: Analysis and model development included the use of 10 378 embryos, all with PGT-A results, from 1385 patients (maternal age range 21-48 years; mean age 36·98 years [SD 4·62]). STORK-A predicted aneuploid versus euploid embryos with an accuracy of 69·3% (95% CI 66·9-71·5; AUC 0·761; positive predictive value [PPV] 76·1%; negative predictive value [NPV] 62·1%) when using images, maternal age, morphokinetics, and blastocyst score. A second classification task trained to predict complex aneuploidy versus euploidy and single aneuploidy produced an accuracy of 74·0% (95% CI 71·7-76·1; AUC 0·760; PPV 54·9%; NPV 87·6%) using an image, maternal age, morphokinetic parameters, and blastocyst grade. A third classification task trained to predict complex aneuploidy versus euploidy had an accuracy of 77·6% (95% CI 75·0-80·0; AUC 0·847; PPV 76·7%; NPV 78·0%). STORK-A reported accuracies of 63·4% (AUC 0·702) on the WCM-ES+ dataset and 65·7% (AUC 0·715) on the IVI Valencia dataset, when using an image, maternal age, and morphokinetic parameters, similar to the STORK-A test dataset accuracy of 67·8% (AUC 0·737), showing generalisability. INTERPRETATION: As a proof of concept, STORK-A shows an ability to predict embryo ploidy in a non-invasive manner and shows future potential as a standardised supplementation to traditional methods of embryo selection and prioritisation for implantation or recommendation for PGT-A. FUNDING: US National Institutes of Health.

Harnessing multimodal data integration to advance precision oncology.

Advances in quantitative biomarker development have accelerated new forms of data-driven insights for patients with cancer. However, most approaches are limited to a single mode of data, leaving integrated approaches across modalities relatively underdeveloped. Multimodal integration of advanced molecular diagnostics, radiological and histological imaging, and codified clinical data presents opportunities to advance precision oncology beyond genomics and standard molecular techniques. However, most medical datasets are still too sparse to be useful for the training of modern machine learning techniques, and significant challenges remain before this is remedied. Combined efforts of data engineering, computational methods for analysis of heterogeneous data and instantiation of synergistic data models in biomedical research are required for success. In this Perspective, we offer our opinions on synthesizing complementary modalities of data with emerging multimodal artificial intelligence methods. Advancing along this direction will result in a reimagined class of multimodal biomarkers to propel the field of precision oncology in the coming decade.

Weakly-supervised tumor purity prediction from frozen H&E stained slides.

BACKGROUND: Estimating tumor purity is especially important in the age of precision medicine. Purity estimates have been shown to be critical for correction of tumor sequencing results, and higher purity samples allow for more accurate interpretations from next-generation sequencing results. Molecular-based purity estimates using computational approaches require sequencing of tumors, which is both time-consuming and expensive. METHODS: Here we propose an approach, weakly-supervised purity (wsPurity), which can accurately quantify tumor purity within a digitally captured hematoxylin and eosin (H&E) stained histological slide, using several types of cancer from The Cancer Genome Atlas (TCGA) as a proof-of-concept. FINDINGS: Our model predicts cancer type with high accuracy on unseen cancer slides from TCGA and shows promising generalizability to unseen data from an external cohort (F1-score of 0.83 for prostate adenocarcinoma). In addition we compare performance of our model on tumor purity prediction with a comparable fully-supervised approach on our TCGA held-out cohort and show our model has improved performance, as well as generalizability to unseen frozen slides (0.1543 MAE on an independent test cohort). In addition to tumor purity prediction, our approach identified high resolution tumor regions within a slide, and can also be used to stratify tumors into high and low tumor purity, using different cancer-dependent thresholds. INTERPRETATION: Overall, we demonstrate our deep learning model's different capabilities to analyze tumor H&E sections. We show our model is generalizable to unseen H&E stained slides from data from TCGA as well as data processed at Weill Cornell Medicine. FUNDING: Starr Cancer Consortium Grant (SCC I15-0027) to Iman Hajirasouliha.

Deep learning predicts chromosomal instability from histopathology images.

Chromosomal instability (CIN) is a hallmark of human cancer yet not readily testable for patients with cancer in routine clinical setting. In this study, we sought to explore whether CIN status can be predicted using ubiquitously available hematoxylin and eosin histology through a deep learning-based model. When applied to a cohort of 1,010 patients with breast cancer (Training set: n = 858, Test set: n = 152) from The Cancer Genome Atlas where 485 patients have high CIN status, our model accurately classified CIN status, achieving an area under the curve of 0.822 with 81.2% sensitivity and 68.7% specificity in the test set. Patch-level predictions of CIN status suggested intra-tumor heterogeneity within slides. Moreover, presence of patches with high predicted CIN score within an entire slide was more predictive of clinical outcome than the average CIN score of the slide, thus underscoring the clinical importance of intra-tumor heterogeneity.

Disruption of Protein Complexes from Weighted Complex Networks.

Essential proteins are indispensable units for living organisms. Removing those leads to disruption of protein complexes and causing lethality. Recently, theoretical methods have been presented to detect essential proteins in protein interaction network. In these methods, an essential protein is predicted as a high-degree vertex of protein interaction network. However, interaction data are usually incomplete and an essential protein cannot have high-connection due to data deficiency. Then, it is critical to design informative networks from other biological data sources. In this paper, we defined a minimal set of proteins to disrupt the maximum number of protein complexes. We constructed a weighted graph using a set of given complexes. We proposed a more appropriate method based on betweenness values to diagnose a minimal set of proteins whose removal would generate the disruption of protein complexes. The effectiveness of the proposed method was benchmarked using given dataset of complexes. The results of our method were compared to the results of other methods in terms of the number of disrupted complexes. Also, results indicated significant superiority of the minimal set of proteins in the massive disruption of complexes. Finally, we investigated the performance of our method for yeast and human datasets and analyzed biological properties of the selected proteins. Our algorithm and some example are freely available from http://bs.ipm.ac.ir/softwares/DPC/DPC.zip.

Deep learning enables robust assessment and selection of human blastocysts after in vitro fertilization.

Visual morphology assessment is routinely used for evaluating of embryo quality and selecting human blastocysts for transfer after in vitro fertilization (IVF). However, the assessment produces different results between embryologists and as a result, the success rate of IVF remains low. To overcome uncertainties in embryo quality, multiple embryos are often implanted resulting in undesired multiple pregnancies and complications. Unlike in other imaging fields, human embryology and IVF have not yet leveraged artificial intelligence (AI) for unbiased, automated embryo assessment. We postulated that an AI approach trained on thousands of embryos can reliably predict embryo quality without human intervention. We implemented an AI approach based on deep neural networks (DNNs) to select highest quality embryos using a large collection of human embryo time-lapse images (about 50,000 images) from a high-volume fertility center in the United States. We developed a framework (STORK) based on Google's Inception model. STORK predicts blastocyst quality with an AUC of >0.98 and generalizes well to images from other clinics outside the US and outperforms individual embryologists. Using clinical data for 2182 embryos, we created a decision tree to integrate embryo quality and patient age to identify scenarios associated with pregnancy likelihood. Our analysis shows that the chance of pregnancy based on individual embryos varies from 13.8% (age ≥41 and poor-quality) to 66.3% (age <37 and good-quality) depending on automated blastocyst quality assessment and patient age. In conclusion, our AI-driven approach provides a reproducible way to assess embryo quality and uncovers new, potentially personalized strategies to select embryos.

Exploring candidate biomarkers for lung and prostate cancers using gene expression and flux variability analysis.

Genome-scale metabolic models have provided valuable resources for exploring changes in metabolism under normal and cancer conditions. However, metabolism itself is strongly linked to gene expression, so integration of gene expression data into metabolic models might improve the detection of genes involved in the control of tumor progression. Herein, we considered gene expression data as extra constraints to enhance the predictive powers of metabolic models. We reconstructed genome-scale metabolic models for lung and prostate, under normal and cancer conditions to detect the major genes associated with critical subsystems during tumor development. Furthermore, we utilized gene expression data in combination with an information theory-based approach to reconstruct co-expression networks of the human lung and prostate in both cohorts. Our results revealed 19 genes as candidate biomarkers for lung and prostate cancer cells. This study also revealed that the development of a complementary approach (integration of gene expression and metabolic profiles) could lead to proposing novel biomarkers and suggesting renovated cancer treatment strategies which have not been possible to detect using either of the methods alone.

Deep Convolutional Neural Networks Enable Discrimination of Heterogeneous Digital Pathology Images.

Pathological evaluation of tumor tissue is pivotal for diagnosis in cancer patients and automated image analysis approaches have great potential to increase precision of diagnosis and help reduce human error. In this study, we utilize several computational methods based on convolutional neural networks (CNN) and build a stand-alone pipeline to effectively classify different histopathology images across different types of cancer. In particular, we demonstrate the utility of our pipeline to discriminate between two subtypes of lung cancer, four biomarkers of bladder cancer, and five biomarkers of breast cancer. In addition, we apply our pipeline to discriminate among four immunohistochemistry (IHC) staining scores of bladder and breast cancers. Our classification pipeline includes a basic CNN architecture, Google's Inceptions with three training strategies, and an ensemble of two state-of-the-art algorithms, Inception and ResNet. Training strategies include training the last layer of Google's Inceptions, training the network from scratch, and fine-tunning the parameters for our data using two pre-trained version of Google's Inception architectures, Inception-V1 and Inception-V3. We demonstrate the power of deep learning approaches for identifying cancer subtypes, and the robustness of Google's Inceptions even in presence of extensive tumor heterogeneity. On average, our pipeline achieved accuracies of 100%, 92%, 95%, and 69% for discrimination of various cancer tissues, subtypes, biomarkers, and scores, respectively. Our pipeline and related documentation is freely available at https://github.com/ih-_lab/CNN_Smoothie.

The Ability of Different Imputation Methods to Preserve the Significant Genes and Pathways in Cancer.

Deciphering important genes and pathways from incomplete gene expression data could facilitate a better understanding of cancer. Different imputation methods can be applied to estimate the missing values. In our study, we evaluated various imputation methods for their performance in preserving significant genes and pathways. In the first step, 5% genes are considered in random for two types of ignorable and non-ignorable missingness mechanisms with various missing rates. Next, 10 well-known imputation methods were applied to the complete datasets. The significance analysis of microarrays (SAM) method was applied to detect the significant genes in rectal and lung cancers to showcase the utility of imputation approaches in preserving significant genes. To determine the impact of different imputation methods on the identification of important genes, the chi-squared test was used to compare the proportions of overlaps between significant genes detected from original data and those detected from the imputed datasets. Additionally, the significant genes are tested for their enrichment in important pathways, using the ConsensusPathDB. Our results showed that almost all the significant genes and pathways of the original dataset can be detected in all imputed datasets, indicating that there is no significant difference in the performance of various imputation methods tested. The source code and selected datasets are available on http://profiles.bs.ipm.ir/softwares/imputation_methods/.

Comparative Analysis of Prostate Cancer Gene Regulatory Networks via Hub Type Variation.

BACKGROUND: Prostate cancer is one of the most widespread cancers in men and is fundamentally a genetic disease. Identifying regulators in cancer using novel systems biology approaches will potentially lead to new insight into this disease. It was sought to address this by inferring gene regulatory networks (GRNs). Moreover, dynamical analysis of GRNs can explain how regulators change among different conditions, such as cancer subtypes. METHODS: In our approach, independent gene regulatory networks from each prostate state were reconstructed using one of the current state-of-art reverse engineering approaches. Next, crucial genes involved in this cancer were highlighted by analyzing each network individually and also in comparison with each other. RESULTS: In this paper, a novel network-based approach was introduced to find critical transcription factors involved in prostate cancer. The results led to detection of 38 essential transcription factors based on hub type variation. Additionally, experimental evidence was found for 29 of them as well as 9 new transcription factors. CONCLUSION: The results showed that dynamical analysis of biological networks may provide useful information to gain better understanding of the cell.

Inferring interaction type in gene regulatory networks using co-expression data.

BACKGROUND: Knowledge of interaction types in biological networks is important for understanding the functional organization of the cell. Currently information-based approaches are widely used for inferring gene regulatory interactions from genomics data, such as gene expression profiles; however, these approaches do not provide evidence about the regulation type (positive or negative sign) of the interaction. RESULTS: This paper describes a novel algorithm, "Signing of Regulatory Networks" (SIREN), which can infer the regulatory type of interactions in a known gene regulatory network (GRN) given corresponding genome-wide gene expression data. To assess our new approach, we applied it to three different benchmark gene regulatory networks, including Escherichia coli, prostate cancer, and an in silico constructed network. Our new method has approximately 68, 70, and 100 percent accuracy, respectively, for these networks. To showcase the utility of SIREN algorithm, we used it to predict previously unknown regulation types for 454 interactions related to the prostate cancer GRN. CONCLUSIONS: SIREN is an efficient algorithm with low computational complexity; hence, it is applicable to large biological networks. It can serve as a complementary approach for a wide range of network reconstruction methods that do not provide information about the interaction type.

Erratum to: Inferring interaction type in gene regulatory networks using co-expression data.

[This corrects the article DOI: 10.1186/s13015-015-0054-4.].

Analysis of candidate genes has proposed the role of y chromosome in human prostate cancer.

BACKGROUND: Prostate cancer, a serious genetic disease, has known as the first widespread cancer in men, but the molecular changes required for the cancer progression has not fully understood. Availability of high-throughput gene expression data has led to the development of various computational methods, for identification of the critical genes, have involved in the cancer. METHODS: In this paper, we have shown the construction of co-expression networks, which have been using Y-chromosome genes, provided an alternative strategy for detecting of new candidate, might involve in prostate cancer. In our approach, we have constructed independent co-expression networks from normal and cancerous stages have been using a reverse engineering approach. Then we have highlighted crucial Y chromosome genes involved in the prostate cancer, by analyzing networks, based on party and date hubs. RESULTS: Our results have led to the detection of 19 critical genes, related to prostate cancer, which 12 of them have previously shown to be involved in this cancer. Also, essential Y chromosome genes have searched based on reconstruction of sub-networks which have led to the identification of 4 experimentally established as well as 4 new Y chromosome genes might be linked putatively to prostate cancer. CONCLUSION: Correct inference of master genes, which mediate molecular, has changed during cancer progression would be one of the major challenges in cancer genomics. In this paper, we have shown the role of Y chromosome genes in finding of the prostate cancer susceptibility genes. Application of our approach to the prostate cancer has led to the establishment of the previous knowledge about this cancer as well as prediction of other new genes.

Network-based approach reveals Y chromosome influences prostate cancer susceptibility.

The human Y chromosome contains a small number of genes that play a critical role in the determination of male-specific organs. Today's advances have provided valuable resources for defining the functions of this chromosome in both normal and cancerous prostates. Despite the fact that generation of high-throughput expression data is becoming usual; the systematic methods of data analysis in a biological context are still an impediment. Here we have shown that constructing co-expression networks using Y-chromosome genes provides an alternative strategy for the detection of new candidate genes involved in prostate cancer. In our approach, independent co-expression networks from normal and cancerous stages are reconstructed using a reverse engineering approach. We then highlight crucial pathways, biological processes, and genes involved in the prostate cancer by analyzing each network individually and in concert. Thus, we have identified 18 critical pathways and processes related to prostate cancer, many of which have previously been shown to be involved in cancer. In particular, we identify 22 Y-chromosome genes putatively linked to prostate cancer, 13 of which have been already verified experimentally. Our novel network-based approach is useful for accurate inference of processes and essential regulators that mediate molecular changes during cancer progression.