Value of machine learning-based transrectal multimodal ultrasound combined with PSA-related indicators in the diagnosis of clinically significant prostate cancer

Zhang, Maoliang; Liu, Yuanzhen; Yao, Jincao; Wang, Kai; Tu, Jing; Hu, Zhengbiao; Jin, Yun; Du, Yue; Sun, Xiang; Chen, Liyu; Wang, Zhengping

doi:10.3389/fendo.2023.1137322

Cited by 3 publications

(1 citation statement)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, studies leveraging these indicators for COVID-19 prediction remain limited. Artificial intelligence (AI) and machine learning are increasingly employed across various fields, with significant innovations in disease prediction, including cardiovascular diseases, neurodegenerative disease [ 5 ], cancer [ 6 ], neurodegenerative diseases [ 7 ], and infectious diseases [ 8 ]. AI can provide relatively accurate and reliable predictions based on diverse data sources and models.…”

Section: Introductionmentioning

confidence: 99%

Online COVID-19 diagnosis prediction using complete blood count: an innovative tool for public health

Teng,

Wang

2023

BMC Public Health

View full text Add to dashboard Cite

Background COVID-19, caused by SARS-CoV-2, presents distinct diagnostic challenges due to its wide range of clinical manifestations and the overlapping symptoms with other common respiratory diseases. This study focuses on addressing these difficulties by employing machine learning (ML) methodologies, particularly the XGBoost algorithm, to utilize Complete Blood Count (CBC) parameters for predictive analysis. Methods We performed a retrospective study involving 2114 COVID-19 patients treated between December 2022 and January 2023 at our healthcare facility. These patients were classified into fever (1057 patients) and pneumonia groups (1057 patients), based on their clinical symptoms. The CBC data were utilized to create predictive models, with model performance evaluated through metrics like Area Under the Receiver Operating Characteristics Curve (AUC), accuracy, sensitivity, specificity, and precision. We selected the top 10 predictive variables based on their significance in disease prediction. The data were then split into a training set (70% of patients) and a validation set (30% of patients) for model validation. Results We identified 31 indicators with significant disparities. The XGBoost model outperformed others, with an AUC of 0.920 and high precision, sensitivity, specificity, and accuracy. The top 10 features (Age, Monocyte%, Mean Platelet Volume, Lymphocyte%, SIRI, Eosinophil count, Platelet count, Hemoglobin, Platelet Distribution Width, and Neutrophil count.) were crucial in constructing a more precise predictive model. The model demonstrated strong performance on both training (AUC = 0.977) and validation (AUC = 0.912) datasets, validated by decision curve analysis and calibration curve. Conclusion ML models that incorporate CBC parameters offer an innovative and effective tool for data analysis in COVID-19. They potentially enhance diagnostic accuracy and the efficacy of therapeutic interventions, ultimately contributing to a reduction in the mortality rate of this infectious disease.

show abstract

Section: Introductionmentioning

confidence: 99%

Online COVID-19 diagnosis prediction using complete blood count: an innovative tool for public health

Teng,

Wang

2023

BMC Public Health

View full text Add to dashboard Cite

show abstract

Using Machine Learning to Select Breast Implant Volume

Basile,

Oliveira

2023

Plastic &Amp; Reconstructive Surgery

View full text Add to dashboard Cite

Background: In breast augmentation surgery, the selection of the appropriate breast implant size is a crucial step that can greatly impact patient satisfaction and the outcome of the procedure. However, this decision is often based on the subjective judgment of the surgeon and the patient, which can lead to suboptimal results. In this study, we aimed to develop a machine-learning approach that can accurately predict the size of breast implants selected for breast augmentation surgery. Methods: We collected data on patient demographics, medical history, and surgeon preferences from a sample of 1000 consecutive patients who underwent breast augmentation. This information was used to train and test a supervised machine learning model to predict the size of breast implants. Results: Our study demonstrated the effectiveness of the algorithm in predicting breast implant size, achieving a Pearson’s correlation coefficient of 0.9335 (p <0.001). The model generated accurate predictions in 86% of the instances, with a Mean Absolute Error (MAE) of 27.10 ml. Its effectiveness was confirmed in the reoperation group, in which 36 of 57 (63%) patients would have received a more suitable implant size if the model's suggestion were followed, potentially avoiding reoperation. Conclusions: Our findings show that machine learning can accurately predict the chosen size of breast implants in augmentation surgery. By integrating our AI model into a decision support system for breast augmentation surgery, essential guidance can be provided to both the surgeons and patients. This not only streamlines the implant selection process but also facilitates enhanced communication and decision-making, ultimately leading to more reliable outcomes and improved patient satisfaction.

show abstract

Radiogenomic analysis of ultrasound phenotypic features coupled to proteomes predicts metastatic risk in primary prostate cancer

Fu,

Luo,

Hong

et al. 2024

BMC Cancer

View full text Add to dashboard Cite

Background Primary prostate cancer with metastasis has a poor prognosis, so assessing its risk of metastasis is essential. Methods This study combined comprehensive ultrasound features with tissue proteomic analysis to obtain biomarkers and practical diagnostic image features that signify prostate cancer metastasis. Results In this study, 17 ultrasound image features of benign prostatic hyperplasia (BPH), primary prostate cancer without metastasis (PPCWOM), and primary prostate cancer with metastasis (PPCWM) were comprehensively analyzed and combined with the corresponding tissue proteome data to perform weighted gene co-expression network analysis (WGCNA), which resulted in two modules highly correlated with the ultrasound phenotype. We screened proteins with temporal expression trends based on the progression of the disease from BPH to PPCWOM and ultimately to PPCWM from two modules and obtained a protein that can promote prostate cancer metastasis. Subsequently, four ultrasound image features significantly associated with the metastatic biomarker HNRNPC (Heterogeneous nuclear ribonucleoprotein C) were identified by analyzing the correlation between the protein and ultrasound image features. The biomarker HNRNPC showed a significant difference in the five-year survival rate of prostate cancer patients (p < 0.0053). On the other hand, we validated the diagnostic efficiency of the four ultrasound image features in clinical data from 112 patients with PPCWOM and 150 patients with PPCWM, obtaining a combined diagnostic AUC of 0.904. In summary, using ultrasound imaging features for predicting whether prostate cancer is metastatic has many applications. Conclusion The above study reveals noninvasive ultrasound image biomarkers and their underlying biological significance, which provide a basis for early diagnosis, treatment, and prognosis of primary prostate cancer with metastasis.

show abstract

Value of machine learning-based transrectal multimodal ultrasound combined with PSA-related indicators in the diagnosis of clinically significant prostate cancer

Cited by 3 publications

References 51 publications

Online COVID-19 diagnosis prediction using complete blood count: an innovative tool for public health

Online COVID-19 diagnosis prediction using complete blood count: an innovative tool for public health

Using Machine Learning to Select Breast Implant Volume

Radiogenomic analysis of ultrasound phenotypic features coupled to proteomes predicts metastatic risk in primary prostate cancer

Contact Info

Product

Resources

About