Prostate cancer prediction using the random forest algorithm that takes into account transrectal ultrasound findings, age, and serum levels of prostate-specific antigen

Li-hong, Xiao; Chen, Pei-Ran; Gou, Zhongping; Li, Yongzhong; Li, Mei; Xiang, Liangcheng; Feng, Ping

doi:10.4103/1008-682x.186884

Cited by 36 publications

(14 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The prostate cancer model showed the highest predictive performance for the optimal models (AUC > 0.80), which is similar to the results obtained in a study using random forest in which an AUC > 0.80 was achieved [ 47 ]. In this study, it was also reported that age was the most contributing variable.…”

Section: Discussionsupporting

confidence: 85%

Prediction of Incident Cancers in the Lifelines Population-Based Cohort

Cortés-Ibañez

Nagaraj

Cornelissen

et al. 2021

Cancers

View full text Add to dashboard Cite

Cancer incidence is rising, and accurate prediction of incident cancers could be relevant to understanding and reducing cancer incidence. The aim of this study was to develop machine learning (ML) models that could predict an incident diagnosis of cancer. Participants without any history of cancer within the Lifelines population-based cohort were followed for a median of 7 years. Data were available for 116,188 cancer-free participants and 4232 incident cancer cases. At baseline, socioeconomic, lifestyle, and clinical variables were assessed. The main outcome was an incident cancer during follow-up (excluding skin cancer), based on linkage with the national pathology registry. The performance of three ML algorithms was evaluated using supervised binary classification to identify incident cancers among participants. Elastic net regularization and Gini index were used for variables selection. An overall area under the receiver operator curve (AUC) <0.75 was obtained, the highest AUC value was for prostate cancer (random forest AUC = 0.82 (95% CI 0.77–0.87), logistic regression AUC = 0.81 (95% CI 0.76–0.86), and support vector machines AUC = 0.83 (95% CI 0.78–0.88), respectively); age was the most important predictor in these models. Linear and non-linear ML algorithms including socioeconomic, lifestyle, and clinical variables produced a moderate predictive performance of incident cancers in the Lifelines cohort.

show abstract

Section: Discussionsupporting

confidence: 85%

Prediction of Incident Cancers in the Lifelines Population-Based Cohort

Cortés-Ibañez

Nagaraj

Cornelissen

et al. 2021

Cancers

View full text Add to dashboard Cite

show abstract

“…RF adopts the concept of integrated learning to synthesize the classification results of each decision tree to prevent over-fitting, thus yielding more accurate and stable results. Xiao et al [19] used the RF method to construct a diagnostic model for prostate cancer with an accuracy of 83.1%, a sensitivity of 65.6%, and a specificity of 93.8%. Casanova et al [20] compared the diagnostic performances of logistic regression and RF for the diagnosis of diabetic retinopathy.…”

Section: Discussionmentioning

confidence: 99%

Identifying tuberculous pleural effusion using artificial intelligence machine learning algorithms

Ren

2019

Respir Res

View full text Add to dashboard Cite

Background The differential diagnosis of tuberculous pleural effusion (TPE) is challenging. In recent years, artificial intelligence (AI) machine learning algorithms have started being used to an increasing extent in disease diagnosis due to the high level of efficiency, objectivity, and accuracy that they offer. Methods Data samples on 192 patients with TPE, 54 patients with parapneumonic pleural effusion (PPE), and 197 patients with malignant pleural effusion (MPE) were retrospectively collected. Based on 28 different features obtained via statistical analysis, TPE diagnostic models using four machine learning algorithms (MLAs), namely logistic regression, k-nearest neighbors (KNN), support vector machine (SVM) and random forest (RF) were established and their respective diagnostic performances were calculated. The respective diagnostic performances of each of the four algorithmic models were compared with that of pleural fluid adenosine deaminase (pfADA). Based on 12 features with the most significant impacts on the accuracy of the RF model, a new RF model was designed for clinical application. To demonstrate its external validity, a prospective study was conducted and the diagnostic performance of the RF model was calculated. Results The respective sensitivity and specificity of each of the four TPE diagnostic models were as follows: logistic regression – 80.5 and 84.8%; KNN– 78.6 and 86.6%; SVM – 83.2 and 85.9%; and RF – 89.1 and 93.6%. The sensitivity and specificity of pfADA were 85.4 and 84.1%, respectively, at the best cut-off value of 17.5 U/L. RF was the superior method among the four MLAs, and was also superior to pfADA. The newly designed RF model (based on 12 out of 28 features) exhibited an acceptable performance rate for the diagnosis of TPE with a sensitivity and specificity of 90.6 and 92.3%, respectively. In the prospective study, its sensitivity and specificity were 100.0 and 90.0%, respectively. Conclusions Establishing a model for the diagnosis of TPE using RF resulted in a more effective, economical, and faster diagnostic method. This method could enable clinicians to diagnose and treat TPE more effectively.

show abstract

“…Our study revealed that sepsis patients with active cancer, when compared to those without active cancer, were predominantly younger in age and were less likely to have chronic illnesses such as diabetes mellitus, hypertension, coronary artery disease, chronic obstructive pulmonary disease and stroke. Although older patients have a higher incidence and a higher prevalence of malignancy in general [ 28 – 32 ], they are also more likely to receive home hospice care compared to younger cancer patients in Taiwan [ 33 ]. We presumed this difference made our cancer patients admitted to ICU younger and had less chronic comorbidities.…”

Section: Discussionmentioning

confidence: 99%

Immune profiles and clinical outcomes between sepsis patients with or without active cancer requiring admission to intensive care units

et al. 2017

View full text Add to dashboard Cite

BackgroundImmunoparalysis was observed in both patients with cancer and sepsis. In cancer patients, Cytotoxic T lymphocyte antigen-4 and programmed cell death protein 1/programmed death-ligand 1 axis are two key components of immunoparalysis. Several emerging therapies against these two axes gained significant clinical benefit. In severe sepsis patients, immunoparalysis was known as compensatory anti-inflammatory response syndrome and this has been suggested as an important cause of death in patients with sepsis. It would be interesting to see if immune status was different in severe sepsis patients with or without active cancer. The aim of this study was to assess the differences in immune profiles, and clinical outcomes between severe sepsis patients with or without cancer admitted to ICU.MethodsA combined retrospective and prospective observational study from a cohort of adult sepsis patients admitted to three medical ICUs at Kaohsiung Chang Gung Memorial Hospital in Taiwan between August 2013 and June 2016.ResultsOf the 2744 patients admitted to the ICU, 532 patients with sepsis were included. Patients were divided into those with or without active cancer according to their medical history. Of the 532 patients, 95 (17.9%) patients had active cancer, and 437 (82.1%) patients had no active cancer history. Patients with active cancer were younger (p = 0.001) and were less likely to have diabetes mellitus (p < 0.001), hypertension (p < 0.001), coronary artery disease (p = 0.004), chronic obstructive pulmonary disease (p = 0.002) or stroke (p = 0.002) compared to patients without active cancer. Patients with active cancer also exhibited higher baseline lactate levels (p = 0.038), and higher baseline plasma interleukin (IL)-10 levels (p = 0.040), higher trend of granulocyte colony-stimulating factor (G-CSF) (p = 0.004) compared to patients without active cancer. The 14-day, 28-day and 90-day mortality rates were higher for patients with active cancer than those without active cancer (P < 0.001 for all intervals).ConclusionsAmong patients admitted to the ICU with sepsis, those with underling active cancer had higher baseline levels of plasma IL-10, higher trend of G-CSF and higher mortality rate than those without active cancer.

show abstract

Prostate cancer prediction using the random forest algorithm that takes into account transrectal ultrasound findings, age, and serum levels of prostate-specific antigen

Cited by 36 publications

References 25 publications

Prediction of Incident Cancers in the Lifelines Population-Based Cohort

Prediction of Incident Cancers in the Lifelines Population-Based Cohort

Identifying tuberculous pleural effusion using artificial intelligence machine learning algorithms

Immune profiles and clinical outcomes between sepsis patients with or without active cancer requiring admission to intensive care units

Contact Info

Product

Resources

About