International Journal of Molecular Sciences

Research

13 pages, 1046 KiB

Open AccessArticle

Applying K-Means Cluster Analysis to Urinary Biomarkers in Interstitial Cystitis/Bladder Pain Syndrome: A New Perspective on Disease Classification

by Yuan-Hong Jiang, Jia-Fong Jhang, Jen-Hung Wang, Ya-Hui Wu and Hann-Chorng Kuo

Int. J. Mol. Sci. 2025, 26(8), 3712; https://doi.org/10.3390/ijms26083712 - 14 Apr 2025

Viewed by 192

Abstract

This study applied K-means cluster analysis to urinary biomarker profiles in interstitial cystitis/bladder pain syndrome (IC/BPS) patients, aiming to provide a new perspective on disease classification and its clinical relevance. We retrospectively analyzed urine samples from 127 IC/BPS patients and 30 controls. The [...] Read more.

This study applied K-means cluster analysis to urinary biomarker profiles in interstitial cystitis/bladder pain syndrome (IC/BPS) patients, aiming to provide a new perspective on disease classification and its clinical relevance. We retrospectively analyzed urine samples from 127 IC/BPS patients and 30 controls. The urinary levels of 10 inflammatory cytokines and three oxidative stress markers (8-hydroxy-2-deoxyguanosin [8-OHdG], 8-isoprostane, and total antioxidant capacity [TAC]) were quantified. K-means clustering was performed to identify biomarker-based patient subgroups. IC/BPS patients exhibited significantly elevated urinary levels of Eotaxin, MCP-1, NGF, 8-OHdG, 8-isoprostane, and TAC compared to controls (all p < 0.05). K-means clustering identified four distinct subgroups. Cluster 4, characterized by the highest levels of inflammatory and oxidative stress biomarkers, comprised 85% ESSIC type 2 IC/BPS patients and exhibited the lowest visual analogue scale (VAS) pain scores and maximal bladder capacity (MBC). Correlation analysis revealed distinct cluster-specific associations between biomarker levels and clinical parameters, including the VAS pain score, MBC, the grade of glomerulation, and treatment outcomes. Applying K-means clustering to urinary inflammatory and oxidative stress biomarkers provides a new perspective on disease classification, identifying IC/BPS subtypes with distinct clinical and biochemical characteristics. This approach may refine disease phenotyping and guide personalized treatment strategies in the future. Full article

(This article belongs to the Special Issue Machine Learning in Disease Diagnosis and Treatment)

► Show Figures

Figure 1

18 pages, 5558 KiB

Open AccessArticle

Deep Clustering-Based Immunotherapy Prediction for Gastric Cancer mRNA Vaccine Development

by Hao Lan, Jinyi Zhao, Linxi Yuan, Menglong Li, Xuemei Pu and Yanzhi Guo

Int. J. Mol. Sci. 2025, 26(6), 2453; https://doi.org/10.3390/ijms26062453 - 10 Mar 2025

Viewed by 582

Abstract

Immunotherapy is becoming a promising strategy for treating diverse cancers. However, it benefits only a selected group of gastric cancer (GC) patients since they have highly heterogeneous immunosuppressive microenvironments. Thus, a more sophisticated immunological subclassification and characterization of GC patients is of great [...] Read more.

Immunotherapy is becoming a promising strategy for treating diverse cancers. However, it benefits only a selected group of gastric cancer (GC) patients since they have highly heterogeneous immunosuppressive microenvironments. Thus, a more sophisticated immunological subclassification and characterization of GC patients is of great practical significance for mRNA vaccine therapy. This study aimed to find a new immunological subclassification for GC and further identify specific tumor antigens for mRNA vaccine development. First, deep autoencoder (AE)-based clustering was utilized to construct the immunological profile and to uncover four distinct immune subtypes of GC, labeled as Subtypes 1, 2, 3, and 4. Then, in silico prediction using machine learning methods was performed for accurate discrimination of new classifications with an average accuracy of 97.6%. Our results suggested significant clinicopathology, molecular, and immune differences across the four subtypes. Notably, Subtype 4 was characterized by poor prognosis, reduced tumor purity, and enhanced immune cell infiltration and activity; thus, tumor-specific antigens associated with Subtype 4 were identified, and a customized mRNA vaccine was developed using immunoinformatic tools. Finally, the influence of the tumor microenvironment (TME) on treatment efficacy was assessed, emphasizing that specific patients may benefit more from this therapeutic approach. Overall, our findings could help to provide new insights into improving the prognosis and immunotherapy of GC patients. Full article

(This article belongs to the Special Issue Machine Learning in Disease Diagnosis and Treatment)

► Show Figures

Figure 1

12 pages, 1439 KiB

Open AccessArticle

GDF15, EGF, and Neopterin in Assessing Progression of Pediatric Chronic Kidney Disease Using Artificial Intelligence Tools—A Pilot Study

by Kinga Musiał, Jakub Stojanowski, Agnieszka Bargenda-Lange and Tomasz Gołębiowski

Int. J. Mol. Sci. 2025, 26(5), 2344; https://doi.org/10.3390/ijms26052344 - 6 Mar 2025

Viewed by 471

Abstract

Cell-mediated immunity and chronic inflammation are hallmarks of chronic kidney disease (CKD). Growth differentiation factor 15 (GDF15) is a marker of inflammation and an integrative signal in stress conditions. Epidermal growth factor (EGF) is a tubule-specific protein that modulates the regeneration of injured [...] Read more.

Cell-mediated immunity and chronic inflammation are hallmarks of chronic kidney disease (CKD). Growth differentiation factor 15 (GDF15) is a marker of inflammation and an integrative signal in stress conditions. Epidermal growth factor (EGF) is a tubule-specific protein that modulates the regeneration of injured renal tubules. Neopterin is a product of activated monocytes and macrophages and serves as a marker of cell-mediated immunity. Our aim was to assess the role of the above-mentioned parameters in the progression of CKD in children using artificial intelligence tools. The study group consisted of 151 children with CKD stages 1–5. EGF, GDF15, and neopterin serum concentrations were assessed by ELISA. The patients’ anthropometric data, biochemical parameters, EGF, GDF15, and neopterin serum values were implemented into the artificial neural network (ANN). The most precise model contained EGF, GDF15, and neopterin as input parameters and classified patients into either CKD 1–3 or CKD 4–5 groups with an excellent accuracy of 96.77%. The presented AI model, with serum concentrations of EGF, GDF15, and neopterin as input parameters, may serve as a useful predictor of CKD progression. It suggests the essential role of inflammatory processes in the renal function decline in the course of CKD in children. Full article

(This article belongs to the Special Issue Machine Learning in Disease Diagnosis and Treatment)

► Show Figures

Figure 1

21 pages, 2185 KiB

Open AccessArticle

Machine Learning Methods for Classifying Multiple Sclerosis and Alzheimer’s Disease Using Genomic Data

by Magdalena Arnal Segura, Giorgio Bini, Anastasia Krithara, Georgios Paliouras and Gian Gaetano Tartaglia

Int. J. Mol. Sci. 2025, 26(5), 2085; https://doi.org/10.3390/ijms26052085 - 27 Feb 2025

Viewed by 644

Abstract

Complex diseases pose challenges in prediction due to their multifactorial and polygenic nature. This study employed machine learning (ML) to analyze genomic data from the UK Biobank, aiming to predict the genomic predisposition to complex diseases like multiple sclerosis (MS) and Alzheimer’s disease [...] Read more.

Complex diseases pose challenges in prediction due to their multifactorial and polygenic nature. This study employed machine learning (ML) to analyze genomic data from the UK Biobank, aiming to predict the genomic predisposition to complex diseases like multiple sclerosis (MS) and Alzheimer’s disease (AD). We tested logistic regression (LR), ensemble tree methods, and deep learning models for this purpose. LR displayed remarkable stability across various subsets of data, outshining deep learning approaches, which showed greater variability in performance. Additionally, ML methods demonstrated an ability to maintain optimal performance despite correlated genomic features due to linkage disequilibrium. When comparing the performance of polygenic risk score (PRS) with ML methods, PRS consistently performed at an average level. By employing explainability tools in the ML models of MS, we found that the results confirmed the polygenicity of this disease. The highest-prioritized genomic variants in MS were identified as expression or splicing quantitative trait loci located in non-coding regions within or near genes associated with the immune response, with a prevalence of human leukocyte antigen (HLA) gene annotations. Our findings shed light on both the potential and the challenges of employing ML to capture complex genomic patterns, paving the way for improved predictive models. Full article

(This article belongs to the Special Issue Machine Learning in Disease Diagnosis and Treatment)

► Show Figures

Figure 1

15 pages, 5464 KiB

Open AccessArticle

Comprehensive Evaluation of Multi-Omics Clustering Algorithms for Cancer Molecular Subtyping

by Juan Wang, Lingxiao Wang, Yi Liu, Xiao Li, Jie Ma, Mansheng Li and Yunping Zhu

Int. J. Mol. Sci. 2025, 26(3), 963; https://doi.org/10.3390/ijms26030963 - 23 Jan 2025

Viewed by 1231

Abstract

As a highly heterogeneous and complex disease, the identification of cancer’s molecular subtypes is crucial for accurate diagnosis and personalized treatment. The integration of multi-omics data enables a comprehensive interpretation of the molecular characteristics of cancer at various biological levels. In recent years, [...] Read more.

As a highly heterogeneous and complex disease, the identification of cancer’s molecular subtypes is crucial for accurate diagnosis and personalized treatment. The integration of multi-omics data enables a comprehensive interpretation of the molecular characteristics of cancer at various biological levels. In recent years, an increasing number of multi-omics clustering algorithms for cancer molecular subtyping have been proposed. However, the absence of a definitive gold standard makes it challenging to evaluate and compare these methods effectively. In this study, we developed a general framework for the comprehensive evaluation of multi-omics clustering algorithms and introduced an innovative metric, the accuracy-weighted average index, which simultaneously considers both clustering performance and clinical relevance. Using this framework, we performed a thorough evaluation and comparison of 11 state-of-the-art multi-omics clustering algorithms, including deep learning-based methods. By integrating the accuracy-weighted average index with computational efficiency, our analysis reveals that PIntMF demonstrates the best overall performance, making it a promising tool for molecular subtyping across a wide range of cancers. Full article

(This article belongs to the Special Issue Machine Learning in Disease Diagnosis and Treatment)

► Show Figures

Figure 1

24 pages, 7359 KiB

Open AccessArticle

Integrative Single-Cell and Bulk RNA Sequencing Identifies a Macrophage-Related Prognostic Signature for Predicting Prognosis and Therapy Responses in Colorectal Cancer

by Shaozhuo Xie, Siyu Hou, Jiajia Chen and Xin Qi

Int. J. Mol. Sci. 2025, 26(2), 811; https://doi.org/10.3390/ijms26020811 - 19 Jan 2025

Viewed by 1172

Abstract

Colorectal cancer (CRC) is one of the most common malignant tumors, characterized by a high incidence and mortality rate. Macrophages, as a key immune cell type within the tumor microenvironment (TME), play a key role in tumor immune evasion and the progression of [...] Read more.

Colorectal cancer (CRC) is one of the most common malignant tumors, characterized by a high incidence and mortality rate. Macrophages, as a key immune cell type within the tumor microenvironment (TME), play a key role in tumor immune evasion and the progression of CRC. Therefore, identifying macrophage biomarkers is of great significance for predicting the prognosis of CRC patients. This study integrates scRNA-seq and bulk RNA-seq data to identify macrophage-related genes in CRC. By applying a comprehensive machine learning framework, the macrophage-related prognostic signature (MRPS) was constructed by 15 macrophage-related genes with prognostic values. The MRPS demonstrated strong predictive performance across multiple datasets, effectively stratifying high-risk and low-risk patients in terms of overall survival (OS) and disease-specific survival (DSS). Furthermore, immune analysis revealed significant differences between the high-risk and low-risk groups in immune cell infiltration levels and immune checkpoint gene expression patterns. Drug screening identified several small molecules, including Bortezomib and Mitoxantrone, as potential therapeutic options for high-risk patients. Pseudotime trajectory analysis further highlighted the potential role of genes comprising the MRPS in macrophage differentiation. This study provides a powerful tool for personalized prognosis prediction in CRC patients, offering new insights into macrophage-driven mechanisms in tumor progression and potential therapeutic strategies. Full article

(This article belongs to the Special Issue Machine Learning in Disease Diagnosis and Treatment)

► Show Figures

Figure 1

9 pages, 2123 KiB

Open AccessArticle

A Decision Tree Model Using Urine Inflammatory and Oxidative Stress Biomarkers for Predicting Lower Urinary Tract Dysfunction in Females

by Yuan-Hong Jiang, Jia-Fong Jhang, Jen-Hung Wang, Ya-Hui Wu and Hann-Chorng Kuo

Int. J. Mol. Sci. 2024, 25(23), 12857; https://doi.org/10.3390/ijms252312857 - 29 Nov 2024

Cited by 1 | Viewed by 823

Abstract

Lower urinary tract dysfunction (LUTD) was associated with bladder inflammation and tissue hypoxia with oxidative stress. The objective of the present study was to investigate the profiles of urine inflammatory and oxidative stress biomarkers in females with LUTD and to develop a urine [...] Read more.

Lower urinary tract dysfunction (LUTD) was associated with bladder inflammation and tissue hypoxia with oxidative stress. The objective of the present study was to investigate the profiles of urine inflammatory and oxidative stress biomarkers in females with LUTD and to develop a urine biomarker-based decision tree model for the prediction. Urine samples were collected from 31 female patients with detrusor overactivity (DO), 45 with dysfunctional voiding (DV), and 114 with bladder pain syndrome (BPS). The targeted analytes included 15 inflammatory cytokines and 3 oxidative stress biomarkers (8-hydroxy-2-deoxyguanosin, 8-isoprostane, and total antioxidant capacity [TAC]). Different female LUTD groups had distinct urine inflammatory and oxidative stress biomarker profiles, including IL-1β, IL-2, IL-8, IL-10, eotaxin, CXCL10, MIP-1β, RANTES, TNFα, VEGF, NGF, BDNF, 8-isoprostane, and TAC. The urine biomarker-based decision tree, using IL-8, IL-10, CXCL10, TNFα, NGF, and BDNF as nodes, demonstrated an overall accuracy rate of 85.3%. The DO, DV, and BPS accuracy rates were 74.2%, 73.3%, and 93.0%, respectively. Internal validation revealed a similar overall accuracy rate. Random forest models supported the significance and importance of all selected nodes in this decision tree model. The inter-individual variations and the presence of extreme values in urine biomarker levels were the limitations of this study. In conclusion, urine inflammatory and oxidative stress biomarker profiles of different female LUTDs were different. This internally validated urine biomarker-based decision tree model predicted different female LUTDs with high accuracy. Full article

(This article belongs to the Special Issue Machine Learning in Disease Diagnosis and Treatment)

► Show Figures

Figure 1

9 pages, 490 KiB

Open AccessCommunication

Advantages of Metabolomics-Based Multivariate Machine Learning to Predict Disease Severity: Example of COVID

by Maryne Lepoittevin, Quentin Blancart Remaury, Nicolas Lévêque, Arnaud W. Thille, Thomas Brunet, Karine Salaun, Mélanie Catroux, Luc Pellerin, Thierry Hauet and Raphael Thuillier

Int. J. Mol. Sci. 2024, 25(22), 12199; https://doi.org/10.3390/ijms252212199 - 13 Nov 2024

Cited by 1 | Viewed by 983

Abstract

The COVID-19 outbreak caused saturations of hospitals, highlighting the importance of early patient triage to optimize resource prioritization. Herein, our objective was to test if high definition metabolomics, combined with ML, can improve prognostication and triage performance over standard clinical parameters using COVID [...] Read more.

The COVID-19 outbreak caused saturations of hospitals, highlighting the importance of early patient triage to optimize resource prioritization. Herein, our objective was to test if high definition metabolomics, combined with ML, can improve prognostication and triage performance over standard clinical parameters using COVID infection as an example. Using high resolution mass spectrometry, we obtained metabolomics profiles of patients and combined them with clinical parameters to design machine learning (ML) algorithms predicting severity (herein determined as the need for mechanical ventilation during patient care). A total of 64 PCR-positive COVID patients at the Poitiers CHU were recruited. Clinical and metabolomics investigations were conducted 8 days after the onset of symptoms. We show that standard clinical parameters could predict severity with good performance (AUC of the ROC curve: 0.85), using SpO2, first respiratory rate, Horowitz quotient and age as the most important variables. However, the performance of the prediction was substantially improved by the use of metabolomics (AUC = 0.92). Our small-scale study demonstrates that metabolomics can improve the performance of diagnosis and prognosis algorithms, and thus be a key player in the future discovery of new biological signals. This technique is easily deployable in the clinic, and combined with machine learning, it can help design the mathematical models needed to advance towards personalized medicine. Full article

(This article belongs to the Special Issue Machine Learning in Disease Diagnosis and Treatment)

► Show Figures

Figure 1

18 pages, 2215 KiB

Open AccessArticle

Mass Spectrometry-Based Metabolomics Reveals a Salivary Signature for Low-Severity COVID-19

by Iasmim Lopes de Lima, Alex Ap. Rosini Silva, Carlos Brites, Natália Angelo da Silva Miyaguti, Felipe Raposo Passos Mansoldo, Sara Vaz Nunes, Pedro Henrique Godoy Sanches, Thais Regiani Cataldi, Caroline Pais de Carvalho, Adriano Reis da Silva, Jonas Ribeiro da Rosa, Mariana Magalhães Borges, Wellisson Vilarindo Oliveira, Thiago Cruz Canevari, Alane Beatriz Vermelho, Marcos Nogueira Eberlin and Andreia M. Porcari

Int. J. Mol. Sci. 2024, 25(22), 11899; https://doi.org/10.3390/ijms252211899 - 6 Nov 2024

Viewed by 1244

Abstract

Omics approaches were extensively applied during the coronavirus disease 2019 (COVID-19) pandemic to understand the disease, identify biomarkers with diagnostic and prognostic value, and discover new molecular targets for medications. COVID-19 continues to challenge the healthcare system as the virus mutates, becoming more [...] Read more.

Omics approaches were extensively applied during the coronavirus disease 2019 (COVID-19) pandemic to understand the disease, identify biomarkers with diagnostic and prognostic value, and discover new molecular targets for medications. COVID-19 continues to challenge the healthcare system as the virus mutates, becoming more transmissible or adept at evading the immune system, causing resurgent epidemic waves over the last few years. In this study, we used saliva from volunteers who were negative and positive for COVID-19 when Omicron and its variants became dominant. We applied a direct solid-phase extraction approach followed by non-target metabolomics analysis to identify potential salivary signatures of hospital-recruited volunteers to establish a model for COVID-19 screening. Our model, which aimed to differentiate COVID-19-positive individuals from controls in a hospital setting, was based on 39 compounds and achieved high sensitivity (85%/100%), specificity (82%/84%), and accuracy (84%/92%) in training and validation sets, respectively. The salivary diagnostic signatures were mainly composed of amino acids and lipids and were related to a heightened innate immune antiviral response and an attenuated inflammatory profile. The higher abundance of thyrotropin-releasing hormone in the COVID-19 positive group highlighted the endocrine imbalance in low-severity disease, as first reported here, underscoring the need for further studies in this area. Full article

(This article belongs to the Special Issue Machine Learning in Disease Diagnosis and Treatment)

► Show Figures

Figure 1

19 pages, 17307 KiB

Open AccessArticle

Targeted Drug Screening Leveraging Senescence-Induced T-Cell Exhaustion Signatures in Hepatocellular Carcinoma

by Qi Qi, Jianyu Pang, Yongzhi Chen, Yuheng Tang, Hui Wang, Samina Gul, Yingjie Sun, Wenru Tang and Miaomiao Sheng

Int. J. Mol. Sci. 2024, 25(20), 11232; https://doi.org/10.3390/ijms252011232 - 18 Oct 2024

Viewed by 1545

Abstract

Hepatocellular carcinoma (HCC) is the sixth most prevalent cancer and a leading cause of cancer-related mortality globally, with most patients diagnosed at advanced stages and facing limited early treatment options. This study aimed to identify characteristic genes associated with T-cell exhaustion due to [...] Read more.

Hepatocellular carcinoma (HCC) is the sixth most prevalent cancer and a leading cause of cancer-related mortality globally, with most patients diagnosed at advanced stages and facing limited early treatment options. This study aimed to identify characteristic genes associated with T-cell exhaustion due to senescence in hepatocellular carcinoma patients, elucidating the interplay between senescence and T-cell exhaustion. We constructed prognostic models based on five signature genes (ENO1, STMN1, PRDX1, RAN, and RANBP1) linked to T-cell exhaustion, utilizing elastic net regression. The findings indicate that increased expression of ENO1 in T cells may contribute to T-cell exhaustion and Treg infiltration in hepatocellular carcinoma. Furthermore, molecular docking was employed to screen small molecule compounds that target the anti-tumor effects of these exhaustion-related genes. This study provides crucial insights into the diagnosis and treatment of hepatocellular carcinoma, establishing a strong foundation for the development of predictive biomarkers and therapeutic targets for affected patients. Full article

(This article belongs to the Special Issue Machine Learning in Disease Diagnosis and Treatment)

► Show Figures

Figure 1

16 pages, 3921 KiB

Open AccessArticle

Predicting Antidiabetic Peptide Activity: A Machine Learning Perspective on Type 1 and Type 2 Diabetes

by Kaida Cai, Zhe Zhang, Wenzhou Zhu, Xiangwei Liu, Tingqing Yu and Wang Liao

Int. J. Mol. Sci. 2024, 25(18), 10020; https://doi.org/10.3390/ijms251810020 - 18 Sep 2024

Viewed by 1711

Abstract

Diabetes mellitus (DM) presents a critical global health challenge, characterized by persistent hyperglycemia and associated with substantial economic and health-related burdens. This study employs advanced machine-learning techniques to improve the prediction and classification of antidiabetic peptides, with a particular focus on differentiating those [...] Read more.

Diabetes mellitus (DM) presents a critical global health challenge, characterized by persistent hyperglycemia and associated with substantial economic and health-related burdens. This study employs advanced machine-learning techniques to improve the prediction and classification of antidiabetic peptides, with a particular focus on differentiating those effective against T1DM from those targeting T2DM. We integrate feature selection with analysis methods, including logistic regression, support vector machines (SVM), and adaptive boosting (AdaBoost), to classify antidiabetic peptides based on key features. Feature selection through the Lasso-penalized method identifies critical peptide characteristics that significantly influence antidiabetic activity, thereby establishing a robust foundation for future peptide design. A comprehensive evaluation of logistic regression, SVM, and AdaBoost shows that AdaBoost consistently outperforms the other methods, making it the most effective approach for classifying antidiabetic peptides. This research underscores the potential of machine learning in the systematic evaluation of bioactive peptides, contributing to the advancement of peptide-based therapies for diabetes management. Full article

(This article belongs to the Special Issue Machine Learning in Disease Diagnosis and Treatment)

► Show Figures

Figure 1

19 pages, 1822 KiB

Open AccessArticle

Renal Cell Carcinoma Discrimination through Attenuated Total Reflection Fourier Transform Infrared Spectroscopy of Dried Human Urine and Machine Learning Techniques

by Bogdan Adrian Buhas, Lucia Ana-Maria Muntean, Guillaume Ploussard, Bogdan Ovidiu Feciche, Iulia Andras, Valentin Toma, Teodor Andrei Maghiar, Nicolae Crișan, Rareș-Ionuț Știufiuc and Constantin Mihai Lucaciu

Int. J. Mol. Sci. 2024, 25(18), 9830; https://doi.org/10.3390/ijms25189830 - 11 Sep 2024

Viewed by 1380

Abstract

Renal cell carcinoma (RCC) is the sixth most common cancer in men and is often asymptomatic, leading to incidental detection in advanced disease stages that are associated with aggressive histology and poorer outcomes. Various cancer biomarkers are found in urine samples from patients [...] Read more.

Renal cell carcinoma (RCC) is the sixth most common cancer in men and is often asymptomatic, leading to incidental detection in advanced disease stages that are associated with aggressive histology and poorer outcomes. Various cancer biomarkers are found in urine samples from patients with RCC. In this study, we propose to investigate the use of Attenuated Total Reflection-Fourier Transform Infrared Spectroscopy (ATR-FTIR) on dried urine samples for distinguishing RCC. We analyzed dried urine samples from 49 patients with RCC, confirmed by histopathology, and 39 healthy donors using ATR-FTIR spectroscopy. The vibrational bands of the dried urine were identified by comparing them with spectra from dried artificial urine, individual urine components, and dried artificial urine spiked with urine components. Urea dominated all spectra, but smaller intensity peaks, corresponding to creatinine, phosphate, and uric acid, were also identified. Statistically significant differences between the FTIR spectra of the two groups were obtained only for creatinine, with lower intensities for RCC cases. The discrimination of RCC was performed through Principal Component Analysis combined with Linear Discriminant Analysis (PCA–LDA) and Support Vector Machine (SVM). Using PCA–LDA, we achieved a higher discrimination accuracy (82%) (using only six Principal Components to avoid overfitting), as compared to SVM (76%). Our results demonstrate the potential of urine ATR-FTIR combined with machine learning techniques for RCC discrimination. However, further studies, especially of other urological diseases, must validate this approach. Full article

(This article belongs to the Special Issue Machine Learning in Disease Diagnosis and Treatment)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Machine Learning in Disease Diagnosis and Treatment

Share This Special Issue

Special Issue Editor

Special Issue Information

Keywords

Benefits of Publishing in a Special Issue

Published Papers (12 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI