Deep learning-based model for predicting progression in patients with head and neck squamous cell carcinoma

A CT-based Deep Learning Radiomics Nomogram for the Prediction of EGFR Mutation Status in Head and Neck Squamous Cell Carcinoma

RATIONALE AND OBJECTIVES

Accurately assessing epidermal growth factor receptor (EGFR) mutation status in head and neck squamous cell carcinoma (HNSCC) patients is crucial for prognosis and treatment selection. This study aimed to construct and validate a contrast-enhanced computed tomography (CECT)-based deep learning radiomics nomogram (DLRN) to predict EGFR mutation status of HNSCC.

MATERIALS AND METHODS

A total of 300 HNSCC patients who underwent CECT scans were enrolled in this study. Participants from two hospitals were separated into a training set (n = 200, 56 EGFR-negative and 144 EGFR-positive) from one hospital and an external test set from the other hospital (n = 100, 37 EGFR-negative and 63 EGFR-positive). The least absolute shrinkage and selection operator method was used to select the key features from CECT-based manually extracted radiomics (MER) features and features automatically extracted using a deep learning model (DL, extracted using a GoogLeNet model). The selected independent clinical factors, MER features, and DL features were then combined to construct a DLRN. The DLRN's performance was evaluated using receiver operating characteristics curves.

RESULTS

Five MER and six DL features were finally chosen. The DLRN, which includes "gender" and "necrotic areas," along with the selected features, predicted EGFR mutation status of HNSCC (EGFR-negative vs. positive) well in both the training (area under the curve [AUC], 0.901) and test (AUC, 0.875) sets.

CONCLUSION

A DLRN using CECT was built to predict EGFR mutation in HNSCC. The model showed high predictive ability and may aid in treatment selection and patient prognosis.

Enhancing head and neck tumor management with artificial intelligence: Integration and perspectives

Head and neck tumors (HNTs) constitute a multifaceted ensemble of pathologies that primarily involve regions such as the oral cavity, pharynx, and nasal cavity. The intricate anatomical structure of these regions poses considerable challenges to efficacious treatment strategies. Despite the availability of myriad treatment modalities, the overall therapeutic efficacy for HNTs continues to remain subdued. In recent years, the deployment of artificial intelligence (AI) in healthcare practices has garnered noteworthy attention. AI modalities, inclusive of machine learning (ML), neural networks (NNs), and deep learning (DL), when amalgamated into the holistic management of HNTs, promise to augment the precision, safety, and efficacy of treatment regimens. The integration of AI within HNT management is intricately intertwined with domains such as medical imaging, bioinformatics, and medical robotics. This article intends to scrutinize the cutting-edge advancements and prospective applications of AI in the realm of HNTs, elucidating AI's indispensable role in prevention, diagnosis, treatment, prognostication, research, and inter-sectoral integration. The overarching objective is to stimulate scholarly discourse and invigorate insights among medical practitioners and researchers to propel further exploration, thereby facilitating superior therapeutic alternatives for patients.

A CT-Based Deep Learning Radiomics Nomogram to Predict Histological Grades of Head and Neck Squamous Cell Carcinoma

RATIONALE AND OBJECTIVES

Accurate pretreatment assessment of histological differentiation grade of head and neck squamous cell carcinoma (HNSCC) is crucial for prognosis evaluation. This study aimed to construct and validate a contrast-enhanced computed tomography (CECT)-based deep learning radiomics nomogram (DLRN) to predict histological differentiation grades of HNSCC.

MATERIALS AND METHODS

A total of 204 patients with HNSCC who underwent CECT scans were enrolled in this study. The participants recruited from two hospitals were split into a training set (n=124, 74 well/moderately differentiated and 50 poorly differentiated) of patients from one hospital and an external test set of patients from the other hospital (n=80, 49 well/moderately differentiated and 31 poorly differentiated). CECT-based manually-extracted radiomics (MER) features and deep learning (DL) features were extracted and selected. The selected MER features and DL features were then combined to construct a DLRN via multivariate logistic regression. The predictive performance of the DLRN was assessed using ROCs and decision curve analysis (DCA).

RESULTS

Three MER features and seven DL features were finally selected. The DLRN incorporating the selected MER and DL features showed good predictive value for the histological differentiation grades of HNSCC (well/moderately differentiated vs. poorly differentiated) in both the training (AUC, 0.878) and test (AUC, 0.822) sets. DCA demonstrated that the DLRN was clinically useful for predicting histological differentiation grades of HNSCC.

CONCLUSION

A CECT-based DLRN was constructed to predict histological differentiation grades of HNSCC. The DLRN showed good predictive efficacy and might be useful for prognostic evaluation of patients with HNSCC.

Subtype-DCC: decoupled contrastive clustering method for cancer subtype identification based on multi-omics data

Due to the high heterogeneity and complexity of cancers, patients with different cancer subtypes often have distinct groups of genomic and clinical characteristics. Therefore, the discovery and identification of cancer subtypes are crucial to cancer diagnosis, prognosis and treatment. Recent technological advances have accelerated the increasing availability of multi-omics data for cancer subtyping. To take advantage of the complementary information from multi-omics data, it is necessary to develop computational models that can represent and integrate different layers of data into a single framework. Here, we propose a decoupled contrastive clustering method (Subtype-DCC) based on multi-omics data integration for clustering to identify cancer subtypes. The idea of contrastive learning is introduced into deep clustering based on deep neural networks to learn clustering-friendly representations. Experimental results demonstrate the superior performance of the proposed Subtype-DCC model in identifying cancer subtypes over the currently available state-of-the-art clustering methods. The strength of Subtype-DCC is also supported by the survival and clinical analysis.

An Unsupervised Deep Learning-Based Model Using Multiomics Data to Predict Prognosis of Patients with Stomach Adenocarcinoma

METHODS

Patients (363 in total) with stomach adenocarcinoma from The Cancer Genome Atlas (TCGA) cohort were included. An autoencoder was constructed to integrate the RNA sequencing, miRNA sequencing, and methylation data. The features of the bottleneck layer were used to perform the k-means clustering algorithm to obtain different subgroups for evaluating the prognosis-related risk of stomach adenocarcinoma. The model's robustness was verified using a 10-fold cross-validation (CV). Survival was analyzed by the Kaplan-Meier method. Univariate and multivariate Cox regression was used to estimate hazard risk. The model was validated in three independent cohorts with different endpoints.

RESULTS

The patients were divided into low-risk and high-risk groups according to the k-means clustering algorithm. The high-risk group had a significantly higher risk of poor survival (log-rank P value = 2.80e - 06; adjusted hazard ratio = 2.386, 95% confidence interval: 1.607~3.543), a concordance index (C-index) of 0.714, and a Brier score of 0.184. The model performed well both in the 10-fold CV procedure and three independent cohorts from the Gene Expression Omnibus (GEO) repository.

CONCLUSIONS

A robust and generalizable model based on the autoencoder was proposed to integrate multiomics data and predict the prognosis of patients with stomach adenocarcinoma. The model demonstrates better performance than two alternative approaches on prognosis prediction. The results might provide the grounds for further exploring the potential biomarkers to predict the prognosis of patients with stomach adenocarcinoma.

Accuracy of artificial intelligence-assisted detection of Oral Squamous Cell Carcinoma: A systematic review and meta-analysis

Oral Squamous Cell Carcinoma (OSCC) is an aggressive tumor with a poor prognosis. Accurate and timely diagnosis is therefore essential for reducing the burden of advanced disease and improving outcomes. In this meta-analysis, we evaluated the accuracy of artificial intelligence (AI)-assisted technologies in detecting OSCC. We included studies that validated any diagnostic modality that used AI to detect OSCC. A search was performed in six databases: PubMed, Embase, Scopus, Cochrane Library, ProQuest, and Web of Science up to 15 Mar 2022. The Quality Assessment Tool for Diagnostic Accuracy Studies was used to evaluate the included studies' quality, while the Split Component Synthesis method was utilized to quantitatively synthesize the pooled diagnostic efficacy estimates. We considered 16 out of the 566 yielded studies, which included twelve different AI models with a total of 6606 samples. The summary sensitivity, summary specificity, positive and negative likelihood ratios as well as the pooled diagnostic odds ratio were 92.0 % (95 % confidence interval [CI] 86.7-95.4 %), 91.9 % (95 % CI 86.5-95.3 %), 11.4 (95 % CI 6.74-19.2), 0.087 (95 % CI 0.051-0.146) and 132 (95 % CI 62.6-277), respectively. Our findings support the capability of AI-assisted systems to detect OSCC with high accuracy, potentially aiding the histopathological examination in early diagnosis, yet more prospective studies are needed to justify their use in the real population.

Combined prognostic value of preoperative serum thyrotrophin and thyroid hormone concentration in papillary thyroid cancer

Background

A growing number of studies have found a close association between thyroid hormones and thyrotrophin (TSH), and they also have prognostic significance in some cancer types; this study aimed to investigate the prognostic value of free triiodothyronine (fT3), free thyroxine (fT4), fT3/fT4, TSH, and their combination in patients with papillary thyroid carcinoma (PTC).

Methods

This study retrospectively analyzed the relevant data of 726 newly diagnosed PTC patients. Both univariate and multivariate analyses were used to predict the recurrence rate, and a risk score was established. In addition, with the use of a random survival forest, a random forest (RF) score was constructed. After calculating the area under the curve (AUC), the diagnostic efficacy of risk score, RF score, and four indicators was compared.

Results

fT3, fT4, fT3/fT4, and TSH were strongly associated with some invasive clinicopathological features and postoperative recurrence. Patients with high expression of fT4 and TSH have a high risk of recurrence. By contrast, patients with high expression of fT3 and fT3/fT4 have a low risk of recurrence. At the same time, the combined use of various indicators is more helpful for establishing an accurate diagnosis. By comparison, we found that the RF score was better than the risk score in terms of predicting the recurrence of PTC.

Conclusion

The diagnostic accuracy of a combination of fT3, fT4, fT3/fT4, and TSH can help improve our clinical estimate of the risk of recurrent PTC, thus allowing the development of a more effective treatment plan for patients.

Deep Learning-Based Multi-Omics Integration Robustly Predicts Relapse in Prostate Cancer

Objective

Post-operative biochemical relapse (BCR) continues to occur in a significant percentage of patients with localized prostate cancer (PCa). Current stratification methods are not adequate to identify high-risk patients. The present study exploits the ability of deep learning (DL) algorithms using the H2O package to combine multi-omics data to resolve this problem.

Methods

Five-omics data from 417 PCa patients from The Cancer Genome Atlas (TCGA) were used to construct the DL-based, relapse-sensitive model. Among them, 265 (63.5%) individuals experienced BCR. Five additional independent validation sets were applied to assess its predictive robustness. Bioinformatics analyses of two relapse-associated subgroups were then performed for identification of differentially expressed genes (DEGs), enriched pathway analysis, copy number analysis and immune cell infiltration analysis.

Results

The DL-based model, with a significant difference (P = 6e-9) between two subgroups and good concordance index (C-index = 0.767), were proven to be robust by external validation. 1530 DEGs including 678 up- and 852 down-regulated genes were identified in the high-risk subgroup S2 compared with the low-risk subgroup S1. Enrichment analyses found five hallmark gene sets were up-regulated while 13 were down-regulated. Then, we found that DNA damage repair pathways were significantly enriched in the S2 subgroup. CNV analysis showed that 30.18% of genes were significantly up-regulated and gene amplification on chromosomes 7 and 8 was significantly elevated in the S2 subgroup. Moreover, enrichment analysis revealed that some DEGs and pathways were associated with immunity. Three tumor-infiltrating immune cell (TIIC) groups with a higher proportion in the S2 subgroup (p = 1e-05, p = 8.7e-06, p = 0.00014) and one TIIC group with a higher proportion in the S1 subgroup (P = 1.3e-06) were identified.

Conclusion

We developed a novel, robust classification for understanding PCa relapse. This study validated the effectiveness of deep learning technique in prognosis prediction, and the method may benefit patients and prevent relapse by improving early detection and advancing early intervention.

Is it useful to use computerized tomography image-based artificial intelligence modelling in the differential diagnosis of chronic otitis media with and without cholesteatoma?

OBJECTIVE

Cholesteatoma is an aggressive form of chronic otitis media (COM). For this reason, it is important to distinguish between COM with and without cholesteatoma. In this study, the role of artificial intelligence modelling in differentiating COM with and without cholesteatoma on computed tomography images was evaluated.

METHODS

The files of 200 patients who underwent mastoidectomy and/or tympanoplasty for COM in our clinic between January 2016 and January 2021 were retrospectively reviewed. According to the presence of cholesteatoma, the patients were divided into two groups as chronic otitis with cholesteatoma (n = 100) and chronic otitis without cholesteatoma (n = 100). The control group (n = 100) consisted of patients who did not have any previous ear disease and did not have any active complaints about the ear. Temporal bone computed tomography (CT) images of all patients were analyzed. The distinction between cholesteatoma and COM was evaluated by using 80% of the CT images obtained for the training of artificial intelligence modelling and the remaining 20% for testing purposes.

RESULTS

The accuracy rate obtained in the hybrid model we used in our study was 95.4%. The proposed model correctly predicted 2952 out of 3093 CT images, while it predicted 141 incorrectly. It correctly predicted 936 (93.78%) of 998 images in the COM group with cholesteatoma, 835 (92.77%) of 900 images in the COM group without cholesteatoma, and 1181 (98.82%) of 1195 images in the normal group.

CONCLUSION

In our study, it has been shown that the differentiation of COM with and without cholesteatoma with artificial intelligence modelling can be made with highly accurate diagnosis rates by using CT images. With the deep learning modelling we proposed, the highest correct diagnosis rate in the literature was obtained. According to the results of our study, we think that with the use of artificial intelligence in practice, the diagnosis of cholesteatoma can be made earlier, it will help in the selection of the most appropriate treatment approach, and the complications can be reduced.

Multimodal deep learning for biomedical data fusion: a review

Biomedical data are becoming increasingly multimodal and thereby capture the underlying complex relationships among biological processes. Deep learning (DL)-based data fusion strategies are a popular approach for modeling these nonlinear relationships. Therefore, we review the current state-of-the-art of such methods and propose a detailed taxonomy that facilitates more informed choices of fusion strategies for biomedical applications, as well as research on novel methods. By doing so, we find that deep fusion strategies often outperform unimodal and shallow approaches. Additionally, the proposed subcategories of fusion strategies show different advantages and drawbacks. The review of current methods has shown that, especially for intermediate fusion strategies, joint representation learning is the preferred approach as it effectively models the complex interactions of different levels of biological organization. Finally, we note that gradual fusion, based on prior biological knowledge or on search strategies, is a promising future research path. Similarly, utilizing transfer learning might overcome sample size limitations of multimodal data sets. As these data sets become increasingly available, multimodal DL approaches present the opportunity to train holistic models that can learn the complex regulatory dynamics behind health and disease.

Molecular Subtypes of Oral Squamous Cell Carcinoma Based on Immunosuppression Genes Using a Deep Learning Approach

Background: The mechanisms through which immunosuppressed patients bear increased risk and worse survival in oral squamous cell carcinoma (OSCC) are unclear. Here, we used deep learning to investigate the genetic mechanisms underlying immunosuppression in the survival of OSCC patients, especially from the aspect of various survival-related subtypes. Materials and methods: OSCC samples data were obtained from The Cancer Genome Atlas (TCGA), International Cancer Genome Consortium (ICGC), and OSCC-related genetic datasets with survival data in the National Center for Biotechnology Information (NCBI). Immunosuppression genes (ISGs) were obtained from the HisgAtlas and DisGeNET databases. Survival analyses were performed to identify the ISGs with significant prognostic values in OSCC. A deep learning (DL)-based model was established for robustly differentiating the survival subpopulations of OSCC samples. In order to understand the characteristics of the different survival-risk subtypes of OSCC samples, differential expression analysis and functional enrichment analysis were performed. Results: A total of 317 OSCC samples were divided into one inferring cohort (TCGA) and four confirmation cohorts (ICGC set, GSE41613, GSE42743, and GSE75538). Eleven ISGs (i.e., BGLAP, CALCA, CTLA4, CXCL8, FGFR3, HPRT1, IL22, ORMDL3, TLR3, SPHK1, and INHBB) showed prognostic value in OSCC. The DL-based model provided two optimal subgroups of TCGA-OSCC samples with significant differences (p = 4.91E-22) and good model fitness [concordance index (C-index) = 0.77]. The DL model was validated by using four external confirmation cohorts: ICGC cohort (n = 40, C-index = 0.39), GSE41613 dataset (n = 97, C-index = 0.86), GSE42743 dataset (n = 71, C-index = 0.87), and GSE75538 dataset (n = 14, C-index = 0.48). Importantly, subtype Sub1 demonstrated a lower probability of survival and thus a more aggressive nature compared with subtype Sub2. ISGs in subtype Sub1 were enriched in the tumor-infiltrating immune cells-related pathways and cancer progression-related pathways, while those in subtype Sub2 were enriched in the metabolism-related pathways. Conclusion: The two survival subtypes of OSCC identified by deep learning can benefit clinical practitioners to divide immunocompromised patients with oral cancer into two subpopulations and give them target drugs and thus might be helpful for improving the survival of these patients and providing novel therapeutic strategies in the precision medicine area.

The landscape of tumors-infiltrate immune cells in papillary thyroid carcinoma and its prognostic value

Introduction

Thyroid cancer is a very common malignant tumor in the endocrine system, while the incidence of papillary thyroid carcinoma (PTC) throughout the world also shows a trend of increase year by year. In this study, we constructed two models: ICIscore and Riskscore. Combined with these two models, we can make more accurate and reasonable inferences about the prognosis of PTC patients.

Methods

We selected 481 PTC samples from TCGA and 147 PTC samples from GEO (49 samples in GSE33630, 65 samples in GSE35570 and 33 samples in GSE60542). We performed consistent clustering for them and divided them into three subgroups and screened differentially expressed genes from these three subgroups. Then we divided the differential genes into three subtypes. We also distinguished the up-regulated and down-regulated genes and calculated ICIscore for each PTC sample. ICIscore consists of two parts: (1) the PCAu was calculated from up-regulated genes. (2) the PCAd was calculated from down-regulated genes. The PCAu and PCAd of each sample were the first principal component of the relevant gene. What’s more, we divided the patients into two groups and constructed mRNA prognostic signatures. Additionally we also verified the independent prognostic value of the signature.

Results

Though ICIscore, we were able to observe the relationship between immune infiltration and prognosis. The result suggests that the activation of the immune system may have both positive and negative consequences. Though Riskscore, we could make more accurate predictions about the prognosis of patients with PTC. Meanwhile, we also generated and validated the ICIscore group and Riskscore group respectively.

Conclusion

All the research results show that by combining the two models constructed, ICIscore and Riskscore, we can make a more accurate and reasonable inference about the prognosis of patients with clinical PTC patients. This suggests that we can provide more effective and reasonable treatment plan for clinical PTC patients.

Deep Learning Reveals Key Immunosuppression Genes and Distinct Immunotypes in Periodontitis

Background

Periodontitis is a chronic immuno-inflammatory disease characterized by inflammatory destruction of tooth-supporting tissues. Its pathogenesis involves a dysregulated local host immune response that is ineffective in combating microbial challenges. An integrated investigation of genes involved in mediating immune response suppression in periodontitis, based on multiple studies, can reveal genes pivotal to periodontitis pathogenesis. Here, we aimed to apply a deep learning (DL)-based autoencoder (AE) for predicting immunosuppression genes involved in periodontitis by integrating multiples omics datasets.

Methods

Two periodontitis-related GEO transcriptomic datasets (GSE16134 and GSE10334) and immunosuppression genes identified from DisGeNET and HisgAtlas were included. Immunosuppression genes related to periodontitis in GSE16134 were used as input to build an AE, to identify the top disease-representative immunosuppression gene features. Using K-means clustering and ANOVA, immune subtype labels were assigned to disease samples and a support vector machine (SVM) classifier was constructed. This classifier was applied to a validation set (Immunosuppression genes related to periodontitis in GSE10334) for predicting sample labels, evaluating the accuracy of the AE. In addition, differentially expressed genes (DEGs), signaling pathways, and transcription factors (TFs) involved in immunosuppression and periodontitis were determined with an array of bioinformatics analysis. Shared DEGs common to DEGs differentiating periodontitis from controls and those differentiating the immune subtypes were considered as the key immunosuppression genes in periodontitis.

Results

We produced representative molecular features and identified two immune subtypes in periodontitis using an AE. Two subtypes were also predicted in the validation set with the SVM classifier. Three “master” immunosuppression genes, PECAM1, FCGR3A, and FOS were identified as candidates pivotal to immunosuppressive mechanisms in periodontitis. Six transcription factors, NFKB1, FOS, JUN, HIF1A, STAT5B, and STAT4, were identified as central to the TFs-DEGs interaction network. The two immune subtypes were distinct in terms of their regulating pathways.

Conclusion

This study applied a DL-based AE for the first time to identify immune subtypes of periodontitis and pivotal immunosuppression genes that discriminated periodontitis from the healthy. Key signaling pathways and TF-target DEGs that putatively mediate immune suppression in periodontitis were identified. PECAM1, FCGR3A, and FOS emerged as high-value biomarkers and candidate therapeutic targets for periodontitis.

Deep Learning in Head and Neck Tumor Multiomics Diagnosis and Analysis: Review of the Literature

Head and neck tumors are the sixth most common neoplasms. Multiomics integrates multiple dimensions of clinical, pathologic, radiological, and biological data and has the potential for tumor diagnosis and analysis. Deep learning (DL), a type of artificial intelligence (AI), is applied in medical image analysis. Among the DL techniques, the convolution neural network (CNN) is used for image segmentation, detection, and classification and in computer-aided diagnosis. Here, we reviewed multiomics image analysis of head and neck tumors using CNN and other DL neural networks. We also evaluated its application in early tumor detection, classification, prognosis/metastasis prediction, and the signing out of the reports. Finally, we highlighted the challenges and potential of these techniques.

Recent Advances in the Application of Artificial Intelligence in Otorhinolaryngology-Head and Neck Surgery

This study presents an up-to-date survey of the use of artificial intelligence (AI) in the field of otorhinolaryngology, considering opportunities, research challenges, and research directions. We searched PubMed, the Cochrane Central Register of Controlled Trials, Embase, and the Web of Science. We initially retrieved 458 articles. The exclusion of non-English publications and duplicates yielded a total of 90 remaining studies. These 90 studies were divided into those analyzing medical images, voice, medical devices, and clinical diagnoses and treatments. Most studies (42.2%, 38/90) used AI for image-based analysis, followed by clinical diagnoses and treatments (24 studies). Each of the remaining two subcategories included 14 studies. Machine learning and deep learning have been extensively applied in the field of otorhinolaryngology. However, the performance of AI models varies and research challenges remain.