RESUMEN
The analysis of super-enhancers (SEs) has recently attracted attention in elucidating the molecular mechanisms of cancer and other diseases. SEs are genomic structures that strongly induce gene expression and have been reported to contribute to the overexpression of oncogenes. Because the analysis of SEs and integrated analysis with other data are performed using large amounts of genome-wide data, artificial intelligence technology, with machine learning at its core, has recently begun to be utilized. In promoting precision medicine, it is important to consider information from SEs in addition to genomic data; therefore, machine learning technology is expected to be introduced appropriately in terms of building a robust analysis platform with a high generalization performance. In this review, we explain the history and principles of SE, and the results of SE analysis using state-of-the-art machine learning and integrated analysis with other data are presented to provide a comprehensive understanding of the current status of SE analysis in the field of medical biology. Additionally, we compared the accuracy between existing machine learning methods on the benchmark dataset and attempted to explore the kind of data preprocessing and integration work needed to make the existing algorithms work on the benchmark dataset. Furthermore, we discuss the issues and future directions of current SE analysis.
Asunto(s)
Algoritmos , Inteligencia Artificial , Aprendizaje Automático , Genómica , Elementos de Facilitación GenéticosRESUMEN
BACKGROUND: In an extensive genomic analysis of lung adenocarcinomas (LUADs), driver mutations have been recognized as potential targets for molecular therapy. However, there remain cases where target genes are not identified. Super-enhancers and structural variants are frequently identified in several hundred loci per case. Despite this, most cancer research has approached the analysis of these data sets separately, without merging and comparing the data, and there are no examples of integrated analysis in LUAD. METHODS: We performed an integrated analysis of super-enhancers and structural variants in a cohort of 174 LUAD cases that lacked clinically actionable genetic alterations. To achieve this, we conducted both WGS and H3K27Ac ChIP-seq analyses using samples with driver gene mutations and those without, allowing for a comprehensive investigation of the potential roles of super-enhancer in LUAD cases. RESULTS: We demonstrate that most genes situated in these overlapped regions were associated with known and previously unknown driver genes and aberrant expression resulting from the formation of super-enhancers accompanied by genomic structural abnormalities. Hi-C and long-read sequencing data further corroborated this insight. When we employed CRISPR-Cas9 to induce structural abnormalities that mimicked cases with outlier ERBB2 gene expression, we observed an elevation in ERBB2 expression. These abnormalities are associated with a higher risk of recurrence after surgery, irrespective of the presence or absence of driver mutations. CONCLUSIONS: Our findings suggest that aberrant gene expression linked to structural polymorphisms can significantly impact personalized cancer treatment by facilitating the identification of driver mutations and prognostic factors, contributing to a more comprehensive understanding of LUAD pathogenesis.
Asunto(s)
Adenocarcinoma del Pulmón , Elementos de Facilitación Genéticos , Regulación Neoplásica de la Expresión Génica , Neoplasias Pulmonares , Receptor ErbB-2 , Humanos , Receptor ErbB-2/genética , Receptor ErbB-2/metabolismo , Adenocarcinoma del Pulmón/genética , Adenocarcinoma del Pulmón/patología , Neoplasias Pulmonares/genética , Neoplasias Pulmonares/patología , Neoplasias Pulmonares/metabolismo , Mutación , Biomarcadores de Tumor/genética , Femenino , Masculino , Variación Estructural del Genoma , Genómica/métodos , Persona de Mediana Edad , Pronóstico , AncianoRESUMEN
The increase in the expectations of artificial intelligence (AI) technology has led to machine learning technology being actively used in the medical field. Non-negative matrix factorization (NMF) is a machine learning technique used for image analysis, speech recognition, and language processing; recently, it is being applied to medical research. Precision medicine, wherein important information is extracted from large-scale medical data to provide optimal medical care for every individual, is considered important in medical policies globally, and the application of machine learning techniques to this end is being handled in several ways. NMF is also introduced differently because of the characteristics of its algorithms. In this review, the importance of NMF in the field of medicine, with a focus on the field of oncology, is described by explaining the mathematical science of NMF and the characteristics of the algorithm, providing examples of how NMF can be used to establish precision medicine, and presenting the challenges of NMF. Finally, the direction regarding the effective use of NMF in the field of oncology is also discussed.
Asunto(s)
Inteligencia Artificial , Medicina de Precisión , Algoritmos , Aprendizaje AutomáticoRESUMEN
BACKGROUND: Wolf-Hirschhorn syndrome candidate gene-1 (WHSC1), a histone methyltransferase, has been found to be upregulated and its expression to be correlated with expression of enhancer of zeste homolog 2 (EZH2) in several cancers. In this study, we evaluated the role of WHSC1 and its therapeutic significance in ovarian clear cell carcinoma (OCCC). METHODS: First, we analyzed WHSC1 expression by quantitative PCR and immunohistochemistry using 23 clinical OCCC specimens. Second, the involvement of WHSC1 in OCCC cell proliferation was evaluated by MTT assays after siRNA-mediated WHSC1 knockdown. We also performed flow cytometry (FACS) to address the effect of WHSC1 on cell cycle. To examine the functional relationship between EZH2 and WHSC1, we knocked down EZH2 using siRNAs and checked the expression levels of WHSC1 and its histone mark H3K36m2 in OCCC cell lines. Finally, we checked WHSC1 expression after treatment with the selective inhibitor, GSK126. RESULTS: Both quantitative PCR and immunohistochemical analysis revealed that WHSC1 was significantly overexpressed in OCCC tissues compared with that in normal ovarian tissues. MTT assay revealed that knockdown of WHSC1 suppressed cell proliferation, and H3K36me2 levels were found to be decreased in immunoblotting. FACS revealed that WHSC1 knockdown affected the cell cycle. We also confirmed that WHSC1 expression was suppressed by EZH2 knockdown or inhibition, indicating that EZH2 is upstream of WHSC1 in OCCC cells. CONCLUSIONS: WHSC1 overexpression induced cell growth and its expression is, at least in part, regulated by EZH2. Further functional analysis will reveal whether WHSC1 is a promising therapeutic target for OCCC.
Asunto(s)
Adenocarcinoma de Células Claras/genética , Proteína Potenciadora del Homólogo Zeste 2/genética , N-Metiltransferasa de Histona-Lisina/genética , Neoplasias Ováricas/genética , Proteínas Represoras/genética , Adenocarcinoma de Células Claras/metabolismo , Línea Celular Tumoral , Proliferación Celular , Proteína Potenciadora del Homólogo Zeste 2/metabolismo , Femenino , Regulación Neoplásica de la Expresión Génica , Técnicas de Silenciamiento del Gen , N-Metiltransferasa de Histona-Lisina/metabolismo , Histonas/metabolismo , Humanos , Neoplasias Ováricas/metabolismo , Proteínas Represoras/metabolismo , Regulación hacia ArribaRESUMEN
DNA methylation is an epigenetic modification that results in dynamic changes during ontogenesis and cell differentiation. DNA methylation patterns regulate gene expression and have been widely researched. While tools for DNA methylation analysis have been developed, most of them have focused on intergroup comparative analysis within a dataset; therefore, it is difficult to conduct cross-dataset studies, such as rare disease studies or cross-institutional studies. This study describes a novel method for DNA methylation analysis, namely, methPLIER, which enables interdataset comparative analyses. methPLIER combines Pathway Level Information Extractor (PLIER), which is a non-negative matrix factorization (NMF) method, with regularization by a knowledge matrix and transfer learning. methPLIER can be used to perform intersample and interdataset comparative analysis based on latent feature matrices, which are obtained via matrix factorization of large-scale data, and factor-loading matrices, which are obtained through matrix factorization of the data to be analyzed. We used methPLIER to analyze a lung cancer dataset and confirmed that the data decomposition reflected sample characteristics for recurrence-free survival. Moreover, methPLIER can analyze data obtained via different preprocessing methods, thereby reducing distributional bias among datasets due to preprocessing. Furthermore, methPLIER can be employed for comparative analyses of methylation data obtained from different platforms, thereby reducing bias in data distribution due to platform differences. methPLIER is expected to facilitate cross-sectional DNA methylation data analysis and enhance DNA methylation data resources.
Asunto(s)
Metilación de ADN , Neoplasias , Humanos , Estudios Transversales , Algoritmos , Epigénesis Genética , Neoplasias/genéticaRESUMEN
High-grade serous ovarian carcinoma (HGSOC) is the most lethal gynecological malignancy. To date, the profiles of gene mutations and copy number alterations in HGSOC have been well characterized. However, the patterns of epigenetic alterations and transcription factor dysregulation in HGSOC have not yet been fully elucidated. In this study, we performed integrative omics analyses of a series of stepwise HGSOC model cells originating from human fallopian tube secretory epithelial cells (HFTSECs) to investigate early epigenetic alterations in HGSOC tumorigenesis. Assay for transposase-accessible chromatin using sequencing (ATAC-seq), chromatin immunoprecipitation sequencing (ChIP-seq), and RNA sequencing (RNA-seq) methods were used to analyze HGSOC samples. Additionally, protein expression changes in target genes were confirmed using normal HFTSECs, serous tubal intraepithelial carcinomas (STICs), and HGSOC tissues. Transcription factor motif analysis revealed that the DNA-binding activity of the AP-1 complex and GATA family proteins was dysregulated during early tumorigenesis. The protein expression levels of JUN and FOSL2 were increased, and those of GATA6 and DAB2 were decreased in STIC lesions, which were associated with epithelial-mesenchymal transition (EMT) and proteasome downregulation. The genomic region around the FRA16D site, containing a cadherin cluster region, was epigenetically suppressed by oncogenic signaling. Proteasome inhibition caused the upregulation of chemokine genes, which may facilitate immune evasion during HGSOC tumorigenesis. Importantly, MEK inhibitor treatment reversed these oncogenic alterations, indicating its clinical effectiveness in a subgroup of patients with HGSOC. This result suggests that MEK inhibitor therapy may be an effective treatment option for chemotherapy-resistant HGSOC.
Asunto(s)
Cistadenocarcinoma Seroso , Neoplasias Ováricas , Femenino , Humanos , Neoplasias Ováricas/metabolismo , Complejo de la Endopetidasa Proteasomal/metabolismo , Cistadenocarcinoma Seroso/genética , Cistadenocarcinoma Seroso/metabolismo , Cistadenocarcinoma Seroso/patología , Carcinogénesis/genética , Factores de Transcripción/metabolismo , Epigénesis Genética , Quinasas de Proteína Quinasa Activadas por Mitógenos/metabolismoRESUMEN
RNA modifications have attracted increasing interest in recent years because they have been frequently implicated in various human diseases, including cancer, highlighting the importance of dynamic posttranscriptional modifications. Methyltransferaselike 6 (METTL6) is a member of the RNA methyltransferase family that has been identified in many cancers; however, little is known about its specific role or mechanism of action. In the present study, we aimed to study the expression levels and functional role of METTL6 in hepatocellular carcinoma (HCC), and further investigate the relevant pathways. To this end, we systematically conducted bioinformatics analysis of METTL6 in HCC using gene expression data and clinical information from a publicly available dataset. The mRNA expression levels of METTL6 were significantly upregulated in HCC tumor tissues compared to that in adjacent nontumor tissues and strongly associated with poorer survival outcomes in patients with HCC. CRISPR/Cas9mediated knockout of METTL6 in HCC cell lines remarkably inhibited colony formation, cell proliferation, cell migration, cell invasion and cell attachment ability. RNA sequencing analysis demonstrated that knockout of METTL6 significantly suppressed the expression of cell adhesionrelated genes. However, chromatin immunoprecipitation sequencing results revealed no significant differences in enhancer activities between cells, which suggests that METTL6 may regulate genes of interest posttranscriptionally. In addition, it was demonstrated for the first time that METTL6 was localized in the cytosol as detected by immunofluorescence analysis, which indicates the plausible location of RNA modification mediated by METTL6. Our findings provide further insight into the function of RNA modifications in cancer and suggest a possible role of METTL6 as a therapeutic target in HCC.
Asunto(s)
Carcinoma Hepatocelular/genética , Moléculas de Adhesión Celular/efectos adversos , ARNt Metiltransferasas/efectos adversos , Carcinoma Hepatocelular/fisiopatología , Moléculas de Adhesión Celular/uso terapéutico , Línea Celular , Movimiento Celular/genética , Movimiento Celular/fisiología , Proliferación Celular/genética , Proliferación Celular/fisiología , Regulación hacia Abajo/genética , Humanos , Neoplasias Hepáticas/genética , Neoplasias Hepáticas/fisiopatología , ARNt Metiltransferasas/metabolismoRESUMEN
High-grade serous ovarian carcinoma (HGSOC) is the most aggressive gynecological malignancy, resulting in approximately 70% of ovarian cancer deaths. However, it is still unclear how genetic dysregulations and biological processes generate the malignant subtype of HGSOC. Here we show that expression levels of microtubule affinity-regulating kinase 3 (MARK3) are downregulated in HGSOC, and that its downregulation significantly correlates with poor prognosis in HGSOC patients. MARK3 overexpression suppresses cell proliferation and angiogenesis of ovarian cancer cells. The LKB1-MARK3 axis is activated by metabolic stress, which leads to the phosphorylation of CDC25B and CDC25C, followed by induction of G2/M phase arrest. RNA-seq and ATAC-seq analyses indicate that MARK3 attenuates cell cycle progression and angiogenesis partly through downregulation of AP-1 and Hippo signaling target genes. The synthetic lethal therapy using metabolic stress inducers may be a promising therapeutic choice to treat the LKB1-MARK3 axis-dysregulated HGSOCs.
Asunto(s)
Quinasas de la Proteína-Quinasa Activada por el AMP/genética , Genes Supresores de Tumor , Neoplasias Ováricas , Proteínas Serina-Treonina Quinasas/genética , Estrés Fisiológico/genética , Biomarcadores de Tumor/genética , Línea Celular Tumoral , Proliferación Celular/genética , Regulación hacia Abajo/genética , Epigénesis Genética/genética , Femenino , Humanos , Neoplasias Ováricas/genética , Neoplasias Ováricas/patologíaRESUMEN
Ovarian clear cell carcinoma (OCCC) has a poor prognosis, and its therapeutic strategy has not been established. PRELP is a leucine-rich repeat protein in the extracellular matrix of connective tissues. Although PRELP anchors the basement membrane to the connective tissue and is absent in most epithelial cancers, much remains unknown regarding its function as a regulator of ligand-mediated signaling pathways. Here, we obtained sets of differentially expressed genes by PRELP expression using OCCC cell lines. We found that more than 1000 genes were significantly altered by PRELP expression, particularly affecting the expression of a group of genes involved in the PI3K-AKT signaling pathway. Furthermore, we revealed the loss of active histone marks on the loci of the PRELP gene in patients with OCCC and how its forced expression inhibited cell proliferation. These findings suggest that PRELP is not only a molecule anchored in connective tissues but is also a signaling molecule acting in a tumor-suppressive manner. It can serve as the basis for early detection and novel therapeutic approaches for OCCC toward precision medicine.
RESUMEN
Since U.S. President Barack Obama announced the Precision Medicine Initiative in his New Year's State of the Union address in 2015, the establishment of a precision medicine system has been emphasized worldwide, particularly in the field of oncology. With the advent of next-generation sequencers specifically, genome analysis technology has made remarkable progress, and there are active efforts to apply genome information to diagnosis and treatment. Generally, in the process of feeding back the results of next-generation sequencing analysis to patients, a molecular tumor board (MTB), consisting of experts in clinical oncology, genetic medicine, etc., is established to discuss the results. On the other hand, an MTB currently involves a large amount of work, with humans searching through vast databases and literature, selecting the best drug candidates, and manually confirming the status of available clinical trials. In addition, as personalized medicine advances, the burden on MTB members is expected to increase in the future. Under these circumstances, introducing cutting-edge artificial intelligence (AI) technology and information and communication technology to MTBs while reducing the burden on MTB members and building a platform that enables more accurate and personalized medical care would be of great benefit to patients. In this review, we introduced the latest status of elemental technologies that have potential for AI utilization in MTB, and discussed issues that may arise in the future as we progress with AI implementation.
RESUMEN
BACKGROUND: Proline/arginine-rich end leucine-rich repeat protein (PRELP) is a member of the small leucine-rich proteoglycan family of extracellular matrix proteins, which is markedly suppressed in the majority of early-stage epithelial cancers and plays a role in regulating the epithelial-mesenchymal transition by altering cell-cell adhesion. Although PRELP is an important factor in the development and progression of bladder cancer, the mechanism of PRELP gene repression remains unclear. RESULTS: Here, we show that repression of PRELP mRNA expression in bladder cancer cells is alleviated by HDAC inhibitors (HDACi) through histone acetylation. Using ChIP-qPCR analysis, we found that acetylation of lysine residue 5 of histone H2B in the PRELP gene promoter region is a marker for the de-repression of PRELP expression. CONCLUSIONS: These results suggest a mechanism through which HDACi may partially regulate the function of PRELP to suppress the development and progression of bladder cancer. Some HDACi are already in clinical use, and the findings of this study provide a mechanistic basis for further investigation of HDACi-based therapeutic strategies.
Asunto(s)
Histonas , Neoplasias de la Vejiga Urinaria , Humanos , Histonas/metabolismo , Inhibidores de Histona Desacetilasas/farmacología , Inhibidores de Histona Desacetilasas/uso terapéutico , Lisina/metabolismo , Glicoproteínas/genética , Acetilación , Neoplasias de la Vejiga Urinaria/tratamiento farmacológico , Neoplasias de la Vejiga Urinaria/genética , Metilación de ADN , Proteínas de la Matriz Extracelular/genética , Proteínas de la Matriz Extracelular/metabolismoRESUMEN
Diagnostic support tools based on artificial intelligence (AI) have exhibited high performance in various medical fields. However, their clinical application remains challenging because of the lack of explanatory power in AI decisions (black box problem), making it difficult to build trust with medical professionals. Nevertheless, visualizing the internal representation of deep neural networks will increase explanatory power and improve the confidence of medical professionals in AI decisions. We propose a novel deep learning-based explainable representation "graph chart diagram" to support fetal cardiac ultrasound screening, which has low detection rates of congenital heart diseases due to the difficulty in mastering the technique. Screening performance improves using this representation from 0.966 to 0.975 for experts, 0.829 to 0.890 for fellows, and 0.616 to 0.748 for residents in the arithmetic mean of area under the curve of a receiver operating characteristic curve. This is the first demonstration wherein examiners used deep learning-based explainable representation to improve the performance of fetal cardiac ultrasound screening, highlighting the potential of explainable AI to augment examiner capabilities.
RESUMEN
In recent years, the diversity of cancer cells in tumor tissues as a result of intratumor heterogeneity has attracted attention. In particular, the development of single-cell analysis technology has made a significant contribution to the field; technologies that are centered on single-cell RNA sequencing (scRNA-seq) have been reported to analyze cancer constituent cells, identify cell groups responsible for therapeutic resistance, and analyze gene signatures of resistant cell groups. However, although single-cell analysis is a powerful tool, various issues have been reported, including batch effects and transcriptional noise due to gene expression variation and mRNA degradation. To overcome these issues, machine learning techniques are currently being introduced for single-cell analysis, and promising results are being reported. In addition, machine learning has also been used in various ways for single-cell analysis, such as single-cell assay of transposase accessible chromatin sequencing (ATAC-seq), chromatin immunoprecipitation sequencing (ChIP-seq) analysis, and multi-omics analysis; thus, it contributes to a deeper understanding of the characteristics of human diseases, especially cancer, and supports clinical applications. In this review, we present a comprehensive introduction to the implementation of machine learning techniques in medical research for single-cell analysis, and discuss their usefulness and future potential.
RESUMEN
Artificial intelligence (AI) is being increasingly adopted in medical research and applications. Medical AI devices have continuously been approved by the Food and Drug Administration in the United States and the responsible institutions of other countries. Ultrasound (US) imaging is commonly used in an extensive range of medical fields. However, AI-based US imaging analysis and its clinical implementation have not progressed steadily compared to other medical imaging modalities. The characteristic issues of US imaging owing to its manual operation and acoustic shadows cause difficulties in image quality control. In this review, we would like to introduce the global trends of medical AI research in US imaging from both clinical and basic perspectives. We also discuss US image preprocessing, ingenious algorithms that are suitable for US imaging analysis, AI explainability for obtaining informed consent, the approval process of medical AI devices, and future perspectives towards the clinical application of AI-based US diagnostic support technologies.
RESUMEN
With the completion of the International Human Genome Project, we have entered what is known as the post-genome era, and efforts to apply genomic information to medicine have become more active. In particular, with the announcement of the Precision Medicine Initiative by U.S. President Barack Obama in his State of the Union address at the beginning of 2015, "precision medicine," which aims to divide patients and potential patients into subgroups with respect to disease susceptibility, has become the focus of worldwide attention. The field of oncology is also actively adopting the precision oncology approach, which is based on molecular profiling, such as genomic information, to select the appropriate treatment. However, the current precision oncology is dominated by a method called targeted-gene panel (TGP), which uses next-generation sequencing (NGS) to analyze a limited number of specific cancer-related genes and suggest optimal treatments, but this method causes the problem that the number of patients who benefit from it is limited. In order to steadily develop precision oncology, it is necessary to integrate and analyze more detailed omics data, such as whole genome data and epigenome data. On the other hand, with the advancement of analysis technologies such as NGS, the amount of data obtained by omics analysis has become enormous, and artificial intelligence (AI) technologies, mainly machine learning (ML) technologies, are being actively used to make more efficient and accurate predictions. In this review, we will focus on whole genome sequencing (WGS) analysis and epigenome analysis, introduce the latest results of omics analysis using ML technologies for the development of precision oncology, and discuss the future prospects.
RESUMEN
In 2019, a novel severe acute respiratory syndrome called coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), was reported and was declared a pandemic by the World Health Organization (WHO) in March 2020. With the advancing development of COVID-19 vaccines and their administration globally, it is expected that COVID-19 will converge in the future; however, the situation remains unpredictable because of a series of reports regarding SARS-CoV-2 variants. Currently, there are still few specific effective treatments for COVID-19, as many unanswered questions remain regarding the pathogenic mechanism of COVID-19. Continued elucidation of COVID-19 pathogenic mechanisms is a matter of global importance. In this regard, recent reports have suggested that epigenetics plays an important role; for instance, the expression of angiotensin I converting enzyme 2 (ACE2) receptor, an important factor in human infection with SARS-CoV-2, is epigenetically regulated; further, DNA methylation status is reported to be unique to patients with COVID-19. In this review, we focus on epigenetic mechanisms to provide a new molecular framework for elucidating the pathogenesis of SARS-CoV-2 infection in humans and of COVID-19, along with the possibility of new diagnostic and therapeutic strategies.
RESUMEN
The coronavirus disease 2019 (COVID-19) pandemic began at the end of December 2019, giving rise to a high rate of infections and causing COVID-19-associated deaths worldwide. It was first reported in Wuhan, China, and since then, not only global leaders, organizations, and pharmaceutical/biotech companies, but also researchers, have directed their efforts toward overcoming this threat. The use of artificial intelligence (AI) has recently surged internationally and has been applied to diverse aspects of many problems. The benefits of using AI are now widely accepted, and many studies have shown great success in medical research on tasks, such as the classification, detection, and prediction of disease, or even patient outcome. In fact, AI technology has been actively employed in various ways in COVID-19 research, and several clinical applications of AI-equipped medical devices for the diagnosis of COVID-19 have already been reported. Hence, in this review, we summarize the latest studies that focus on medical imaging analysis, drug discovery, and therapeutics such as vaccine development and public health decision-making using AI. This survey clarifies the advantages of using AI in the fight against COVID-19 and provides future directions for tackling the COVID-19 pandemic using AI techniques.
RESUMEN
Although chromatin immunoprecipitation and next-generation sequencing (ChIP-seq) using formalin-fixed paraffin-embedded tissue (FFPE) has been reported, it remained elusive whether they retained accurate transcription factor binding. Here, we developed a method to identify the binding sites of the insulator transcription factor CTCF and the genome-wide distribution of histone modifications involved in transcriptional activation. Importantly, we provide evidence that the ChIP-seq datasets obtained from FFPE samples are similar to or even better than the data for corresponding fresh-frozen samples, indicating that FFPE samples are compatible with ChIP-seq analysis. H3K27ac ChIP-seq analyses of 69 FFPE samples using a dual-arm robot revealed that driver mutations in EGFR were distinguishable from pan-negative cases and were relatively homogeneous as a group in lung adenocarcinomas. Thus, our results demonstrate that FFPE samples are an important source for epigenomic research, enabling the study of histone modifications, nuclear chromatin structure, and clinical data.
RESUMEN
Image segmentation is the pixel-by-pixel detection of objects, which is the most challenging but informative in the fundamental tasks of machine learning including image classification and object detection. Pixel-by-pixel segmentation is required to apply machine learning to support fetal cardiac ultrasound screening; we have to detect cardiac substructures precisely which are small and change shapes dynamically with fetal heartbeats, such as the ventricular septum. This task is difficult for general segmentation methods such as DeepLab v3+, and U-net. Hence, here we proposed a novel segmentation method named Cropping-Segmentation-Calibration (CSC) that is specific to the ventricular septum in ultrasound videos in this study. CSC employs the time-series information of videos and specific section information to calibrate the output of U-net. The actual sections of the ventricular septum were annotated in 615 frames from 421 normal fetal cardiac ultrasound videos of 211 pregnant women who were screened. The dataset was assigned a ratio of 2:1, which corresponded to a ratio of the training to test data, and three-fold cross-validation was conducted. The segmentation results of DeepLab v3+, U-net, and CSC were evaluated using the values of the mean intersection over union (mIoU), which were 0.0224, 0.1519, and 0.5543, respectively. The results reveal the superior performance of CSC.
Asunto(s)
Aprendizaje Profundo , Feto/diagnóstico por imagen , Procesamiento de Imagen Asistido por Computador/métodos , Tabique Interventricular/diagnóstico por imagen , Femenino , Humanos , Embarazo , Factores de Tiempo , UltrasonografíaRESUMEN
The application of segmentation methods to medical imaging has the potential to create novel diagnostic support models. With respect to fetal ultrasound, the thoracic wall is a key structure on the assessment of the chest region for examiners to recognize the relative orientation and size of structures inside the thorax, which are critical components in neonatal prognosis. In this study, to improve the segmentation performance of the thoracic wall in fetal ultrasound videos, we proposed a novel model-agnostic method using deep learning techniques: the Multi-Frame + Cylinder method (MFCY). The Multi-frame method (MF) uses time-series information of ultrasound videos, and the Cylinder method (CY) utilizes the shape of the thoracic wall. To evaluate the achieved improvement, we performed segmentation using five-fold cross-validation on 538 ultrasound frames in the four-chamber view (4CV) of 256 normal cases using U-net and DeepLabv3+. MFCY increased the mean values of the intersection over union (IoU) of thoracic wall segmentation from 0.448 to 0.493 for U-net and from 0.417 to 0.470 for DeepLabv3+. These results demonstrated that MFCY improved the segmentation performance of the thoracic wall in fetal ultrasound videos without altering the network structure. MFCY is expected to facilitate the development of diagnostic support models in fetal ultrasound by providing further accurate segmentation of the thoracic wall.