Diagnostics

Research

Jump to: Other

23 pages, 3524 KiB

Open AccessArticle

Predicting Sudden Sensorineural Hearing Loss Recovery with Patient-Personalized Seigel’s Criteria Using Machine Learning

by Sanghyun Shon, Kanghyeon Lim, Minsu Chae, Hwamin Lee and June Choi

Diagnostics 2024, 14(12), 1296; https://doi.org/10.3390/diagnostics14121296 - 19 Jun 2024

Viewed by 1411

Abstract

Background: Accurate prognostic prediction is crucial for managing Idiopathic Sudden Sensorineural Hearing Loss (ISSHL). Previous studies developing ISSHL prognosis models often overlooked individual variability in hearing damage by relying on fixed frequency domains. This study aims to develop models predicting ISSHL prognosis one [...] Read more.

Background: Accurate prognostic prediction is crucial for managing Idiopathic Sudden Sensorineural Hearing Loss (ISSHL). Previous studies developing ISSHL prognosis models often overlooked individual variability in hearing damage by relying on fixed frequency domains. This study aims to develop models predicting ISSHL prognosis one month after treatment, focusing on patient-specific hearing impairments. Methods: Patient-Personalized Seigel’s Criteria (PPSC) were developed considering patient-specific hearing impairment related to ISSHL criteria. We performed a statistical test to assess the shift in the recovery assessment when applying PPSC. The utilized dataset of 581 patients comprised demographic information, health records, laboratory testing, onset and treatment, and hearing levels. To reduce the model’s reliance on hearing level features, we used only the averages of hearing levels of the impaired frequencies. Then, model development, evaluation, and interpretation proceeded. Results: The chi-square test (p-value: 0.106) indicated that the shift in recovery assessment is not statistically significant. The soft-voting ensemble model was most effective, achieving an Area Under the Receiver Operating Characteristic Curve (AUROC) of 0.864 (95% CI: 0.801–0.927), with model interpretation based on the SHapley Additive exPlanations value. Conclusions: With PPSC, providing a hearing assessment comparable to traditional Seigel’s criteria, the developed models successfully predicted ISSHL recovery one month post-treatment by considering patient-specific impairments. Full article

(This article belongs to the Special Issue Machine Learning in Precise and Personalized Diagnosis)

► Show Figures

Figure 1

20 pages, 27178 KiB

Open AccessArticle

Classification and Determination of Severity of Corneal Ulcer with Vision Transformer Based on the Analysis of Public Image Dataset of Fluorescein-Stained Corneas

by Talha Burak Alakuş and Muhammet Baykara

Diagnostics 2024, 14(8), 786; https://doi.org/10.3390/diagnostics14080786 - 9 Apr 2024

Cited by 1 | Viewed by 1743

Abstract

A corneal ulcer is a condition in which an injury to the corneal surface occurs as a result of infection. This can lead to severe vision loss and even blindness. For this reason, early diagnosis of this disease is of great importance. Deep [...] Read more.

A corneal ulcer is a condition in which an injury to the corneal surface occurs as a result of infection. This can lead to severe vision loss and even blindness. For this reason, early diagnosis of this disease is of great importance. Deep learning algorithms are used in many critical health applications and are used effectively in the early diagnosis stages of diseases. Thus, a deep learning algorithm was applied in this study and corneal ulcer and severity were predicted. The study consisted of four stages over three different scenarios. In the first scenario, the types of corneal ulcers were predicted. In the second scenario, the grades of corneal ulcer types were classified. In the last scenario, the severity of corneal ulcers was classified. For each scenario, data were obtained in the first stage and separated according to the relevant labels. In the second stage, various image processing algorithms were employed, and images were analyzed. At this stage, the images were also augmented by various processes. In the third stage, ViT architecture, a new deep learning model, was used, and the images were classified. In the last stage, the performance of the classifier was determined by accuracy, precision, recall, F1-score, and AUC score. At the end of the study, the ViT deep learning model performed an effective classification, and accuracy scores of 95.77% for the first scenario, 96.43% for the second scenario, and 97.27% for the third scenario were calculated. Full article

(This article belongs to the Special Issue Machine Learning in Precise and Personalized Diagnosis)

► Show Figures

Figure 1

21 pages, 9700 KiB

Open AccessArticle

Crafting a Personalized Prognostic Model for Malignant Prostate Cancer Patients Using Risk Gene Signatures Discovered through TCGA-PRAD Mining, Machine Learning, and Single-Cell RNA-Sequencing

by Feng Lyu, Xianshu Gao, Mingwei Ma, Mu Xie, Shiyu Shang, Xueying Ren, Mingzhu Liu and Jiayan Chen

Diagnostics 2023, 13(12), 1997; https://doi.org/10.3390/diagnostics13121997 - 7 Jun 2023

Cited by 3 | Viewed by 2456

Abstract

Background: Prostate cancer is a significant clinical issue, particularly for high Gleason score (GS) malignancy patients. Our study aimed to engineer and validate a risk model based on the profiles of high-GS PCa patients for early identification and the prediction of prognosis. Methods: [...] Read more.

Background: Prostate cancer is a significant clinical issue, particularly for high Gleason score (GS) malignancy patients. Our study aimed to engineer and validate a risk model based on the profiles of high-GS PCa patients for early identification and the prediction of prognosis. Methods: We conducted differential gene expression analysis on patient samples from The Cancer Genome Atlas (TCGA) and enriched our understanding of gene functions. Using the least absolute selection and shrinkage operator (LASSO) regression, we established a risk model and validated it using an independent dataset from the International Cancer Genome Consortium (ICGC). Clinical variables were incorporated into a nomogram to predict overall survival (OS), and machine learning was used to explore the risk factor characteristics’ impact on PCa prognosis. Our prognostic model was confirmed using various databases, including single-cell RNA-sequencing datasets (scRNA-seq), the Cancer Cell Line Encyclopedia (CCLE), PCa cell lines, and tumor tissues. Results: We identified 83 differentially expressed genes (DEGs). Furthermore, WASIR1, KRTAP5-1, TLX1, KIF4A, and IQGAP3 were determined to be significant risk factors for OS and progression-free survival (PFS). Based on these five risk factors, we developed a risk model and nomogram for predicting OS and PFS, with a C-index of 0.823 (95% CI, 0.766–0.881) and a 10-year area under the curve (AUC) value of 0.788 (95% CI, 0.633–0.943). Additionally, the 3-year AUC was 0.759 when validating using ICGC. KRTAP5-1 and WASIR1 were found to be the most influential prognosis factors when using the optimized machine learning model. Finally, the established model was interrelated with immune cell infiltration, and the signals were found to be differentially expressed in PCa cells when using scRNA-seq datasets and tissues. Conclusions: We engineered an original and novel prognostic model based on five gene signatures through TCGA and machine learning, providing new insights into the risk of scarification and survival prediction for PCa patients in clinical practice. Full article

(This article belongs to the Special Issue Machine Learning in Precise and Personalized Diagnosis)

► Show Figures

Figure 1

17 pages, 2136 KiB

Open AccessArticle

An Invasive Ductal Carcinomas Breast Cancer Grade Classification Using an Ensemble of Convolutional Neural Networks

by Eelandula Kumaraswamy, Sumit Kumar and Manoj Sharma

Diagnostics 2023, 13(11), 1977; https://doi.org/10.3390/diagnostics13111977 - 5 Jun 2023

Cited by 14 | Viewed by 2590

Abstract

Invasive Ductal Carcinoma Breast Cancer (IDC-BC) is the most common type of cancer and its asymptomatic nature has led to an increased mortality rate globally. Advancements in artificial intelligence and machine learning have revolutionized the medical field with the development of AI-enabled computer-aided [...] Read more.

Invasive Ductal Carcinoma Breast Cancer (IDC-BC) is the most common type of cancer and its asymptomatic nature has led to an increased mortality rate globally. Advancements in artificial intelligence and machine learning have revolutionized the medical field with the development of AI-enabled computer-aided diagnosis (CAD) systems, which help in determining diseases at an early stage. CAD systems assist pathologists in their decision-making process to produce more reliable outcomes in order to treat patients well. In this work, the potential of pre-trained convolutional neural networks (CNNs) (i.e., EfficientNetV2L, ResNet152V2, DenseNet201), singly or as an ensemble, was thoroughly explored. The performances of these models were evaluated for IDC-BC grade classification using the DataBiox dataset. Data augmentation was used to avoid the issues of data scarcity and data imbalances. The performance of the best model was compared to three different balanced datasets of Databiox (i.e., 1200, 1400, and 1600 images) to determine the implications of this data augmentation. Furthermore, the effects of the number of epochs were analysed to ensure the coherency of the most optimal model. The experimental results analysis revealed that the proposed ensemble model outperformed the existing state-of-the-art techniques in relation to classifying the IDC-BC grades of the Databiox dataset. The proposed ensemble model of the CNNs achieved a 94% classification accuracy and attained a significant area under the ROC curves for grades 1, 2, and 3, i.e., 96%, 94%, and 96%, respectively. Full article

(This article belongs to the Special Issue Machine Learning in Precise and Personalized Diagnosis)

► Show Figures

Figure 1

17 pages, 19336 KiB

Open AccessArticle

Artificial-Intelligence-Driven Algorithms for Predicting Response to Corticosteroid Treatment in Patients with Post-Acute COVID-19

by Vojtech Myska, Samuel Genzor, Anzhelika Mezina, Radim Burget, Jan Mizera, Michal Stybnar, Martin Kolarik, Milan Sova and Malay Kishore Dutta

Diagnostics 2023, 13(10), 1755; https://doi.org/10.3390/diagnostics13101755 - 16 May 2023

Cited by 5 | Viewed by 2312

Abstract

Pulmonary fibrosis is one of the most severe long-term consequences of COVID-19. Corticosteroid treatment increases the chances of recovery; unfortunately, it can also have side effects. Therefore, we aimed to develop prediction models for a personalized selection of patients benefiting from corticotherapy. The [...] Read more.

Pulmonary fibrosis is one of the most severe long-term consequences of COVID-19. Corticosteroid treatment increases the chances of recovery; unfortunately, it can also have side effects. Therefore, we aimed to develop prediction models for a personalized selection of patients benefiting from corticotherapy. The experiment utilized various algorithms, including Logistic Regression, k-NN, Decision Tree, XGBoost, Random Forest, SVM, MLP, AdaBoost, and LGBM. In addition easily human-interpretable model is presented. All algorithms were trained on a dataset consisting of a total of 281 patients. Every patient conducted an examination at the start and three months after the post-COVID treatment. The examination comprised a physical examination, blood tests, functional lung tests, and an assessment of health state based on X-ray and HRCT. The Decision tree algorithm achieved balanced accuracy (BA) of 73.52%, ROC-AUC of 74.69%, and 71.70% F1 score. Other algorithms achieving high accuracy included Random Forest (BA 70.00%, ROC-AUC 70.62%, 67.92% F1 score) and AdaBoost (BA 70.37%, ROC-AUC 63.58%, 70.18% F1 score). The experiments prove that information obtained during the initiation of the post-COVID-19 treatment can be used to predict whether the patient will benefit from corticotherapy. The presented predictive models can be used by clinicians to make personalized treatment decisions. Full article

(This article belongs to the Special Issue Machine Learning in Precise and Personalized Diagnosis)

► Show Figures

Figure 1

14 pages, 2657 KiB

Open AccessArticle

Automated Detection of Endometrial Polyps from Hysteroscopic Videos Using Deep Learning

by Aihua Zhao, Xin Du, Suzhen Yuan, Wenfeng Shen, Xin Zhu and Wenwen Wang

Diagnostics 2023, 13(8), 1409; https://doi.org/10.3390/diagnostics13081409 - 13 Apr 2023

Cited by 5 | Viewed by 6191

Abstract

Endometrial polyps are common gynecological lesions. The standard treatment for this condition is hysteroscopic polypectomy. However, this procedure may be accompanied by misdetection of endometrial polyps. To improve the diagnostic accuracy and reduce the risk of misdetection, a deep learning model based on [...] Read more.

Endometrial polyps are common gynecological lesions. The standard treatment for this condition is hysteroscopic polypectomy. However, this procedure may be accompanied by misdetection of endometrial polyps. To improve the diagnostic accuracy and reduce the risk of misdetection, a deep learning model based on YOLOX is proposed to detect endometrial polyps in real time. Group normalization is employed to improve its performance with large hysteroscopic images. In addition, we propose a video adjacent-frame association algorithm to address the problem of unstable polyp detection. Our proposed model was trained on a dataset of 11,839 images from 323 cases provided by a hospital and was tested on two datasets of 431 cases from two hospitals. The results show that the lesion-based sensitivity of the model reached 100% and 92.0% for the two test sets, compared with 95.83% and 77.33%, respectively, for the original YOLOX model. This demonstrates that the improved model may be used effectively as a diagnostic tool during clinical hysteroscopic procedures to reduce the risk of missing endometrial polyps. Full article

(This article belongs to the Special Issue Machine Learning in Precise and Personalized Diagnosis)

► Show Figures

Figure 1

21 pages, 3509 KiB

Open AccessArticle

Construct ceRNA Network and Risk Model of Breast Cancer Using Machine Learning Methods under the Mechanism of Cuproptosis

by Jianzhi Deng, Fei Fu, Fengming Zhang, Yuanyuan Xia and Yuehan Zhou

Diagnostics 2023, 13(6), 1203; https://doi.org/10.3390/diagnostics13061203 - 22 Mar 2023

Cited by 4 | Viewed by 2693

Abstract

Breast cancer (BRCA) has an undesirable prognosis and is the second most common cancer among women after lung cancer. A novel mechanism of programmed cell death called cuproptosis is linked to the development and spread of tumor cells. However, the function of cuproptosis [...] Read more.

Breast cancer (BRCA) has an undesirable prognosis and is the second most common cancer among women after lung cancer. A novel mechanism of programmed cell death called cuproptosis is linked to the development and spread of tumor cells. However, the function of cuproptosis in BRCA remains unknown. To this date, no studies have used machine learning methods to screen for characteristic genes to explore the role of cuproptosis-related genes (CRGs) in breast cancer. Therefore, 14 cuproptosis-related characteristic genes (CRCGs) were discovered by the feature selection of 39 differentially expressed CRGs using the three machine learning methods LASSO, SVM-RFE, and random forest. Through the PPI network and immune infiltration analysis, we found that PRNP was the key CRCG. The miRTarBase, TargetScan, and miRDB databases were then used to identify hsa-miR-192-5p and hsa-miR-215-5p as the upstream miRNA of PRNP, and the upstream lncRNA, CARMN, was identified by the StarBase database. Thus, the mRNA PRNP/miRNA hsa-miR-192-5p and hsa-miR-215-5p/lncRNA CARMN ceRNA network was constructed. This ceRNA network, which has not been studied before, is extremely innovative. Furthermore, four cuproptosis-related lncRNAs (CRLs) were screened in TCGA-BRCA by univariate Cox, LASSO, and multivariate Cox regression analysis. The risk model was constructed by using these four CRLs, and the risk score = C9orf163 * (1.8365) + PHC2-AS1 * (−2.2985) + AC087741.1 * (−0.9504) + AL109824.1 * (0.6016). The ROC curve and C-index demonstrated the superior predictive capacity of the risk model, and the ROC curve demonstrated that the AUC of 1-, 3-, and 5-year OS in all samples was 0.721, 0.695, and 0.633, respectively. Finally, 50 prospective sensitive medicines were screened with the pRRophetic R package, among which 17-AAG may be a therapeutic agent for high-risk patients, while the other 49 medicines may be suitable for the treatment of low-risk patients. In conclusion, our study constructs a new ceRNA network and a novel risk model, which offer a theoretical foundation for the treatment of BRCA and will aid in improving the prognosis of BRCA. Full article

(This article belongs to the Special Issue Machine Learning in Precise and Personalized Diagnosis)

► Show Figures

Figure 1

Other

Jump to: Research

8 pages, 1165 KiB

Open AccessBrief Report

A New Regression Model for Depression Severity Prediction Based on Correlation among Audio Features Using a Graph Convolutional Neural Network

by Momoko Ishimaru, Yoshifumi Okada, Ryunosuke Uchiyama, Ryo Horiguchi and Itsuki Toyoshima

Diagnostics 2023, 13(4), 727; https://doi.org/10.3390/diagnostics13040727 - 14 Feb 2023

Cited by 3 | Viewed by 2002

Abstract

Recent studies have revealed mutually correlated audio features in the voices of depressed patients. Thus, the voices of these patients can be characterized based on the combinatorial relationships among the audio features. To date, many deep learning–based methods have been proposed to predict [...] Read more.

Recent studies have revealed mutually correlated audio features in the voices of depressed patients. Thus, the voices of these patients can be characterized based on the combinatorial relationships among the audio features. To date, many deep learning–based methods have been proposed to predict the depression severity using audio data. However, existing methods have assumed that the individual audio features are independent. Hence, in this paper, we propose a new deep learning–based regression model that allows for the prediction of depression severity on the basis of the correlation among audio features. The proposed model was developed using a graph convolutional neural network. This model trains the voice characteristics using graph-structured data generated to express the correlation among audio features. We conducted prediction experiments on depression severity using the DAIC-WOZ dataset employed in several previous studies. The experimental results showed that the proposed model achieved a root mean square error (RMSE) of 2.15, a mean absolute error (MAE) of 1.25, and a symmetric mean absolute percentage error of 50.96%. Notably, RMSE and MAE significantly outperformed the existing state-of-the-art prediction methods. From these results, we conclude that the proposed model can be a promising tool for depression diagnosis. Full article

(This article belongs to the Special Issue Machine Learning in Precise and Personalized Diagnosis)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Machine Learning in Precise and Personalized Diagnosis

Share This Special Issue

Special Issue Editor

Special Issue Information

Benefits of Publishing in a Special Issue

Published Papers (8 papers)

Research

Other

Further Information

Guidelines

MDPI Initiatives

Follow MDPI