Logistic Regression Algorithm Research Articles

A novel ensemble approach for Twitter sentiment classification with ML and LSTM algorithms for real-time tweets analysis

Social media sentiment classification was an essential consideration in natural language processing (NLP) for evaluating normal people’s perspectives on a given topic. With Twitter’s massive rise in popularity in recent years, the capacity to extract information about public sentiment from tweets became a major focus. This paper not only analyzed public sentiment through data from Twitter but introduced a novel ensemble approach in the methods employed for Twitter sentiment classification. Real-time tweets on various topics, including “covid,” “crime,” “spam,” “flipkart,” “migraine,” and “airlines,” were extracted and thoroughly examined to gain insight into public opinions. Leveraging the Twitter API for real-time tweet extraction, natural language processing techniques were applied to clean the tweet data. Subsequently, we applied several machine learning (ML) algorithms Naïve Bayes, decision tree (DT), random forest (RF), logistic regression (LGR), and deep learning (DL) algorithms recurrent neural network (RNN), LSTM, and GRU individually. Later, we proposed a novel ensemble of ML and DL algorithms for sentiment classification, with a novel emphasis on ensemble techniques and enhanced the accuracy with a significance compared to individual ML or DL model applied. The experimental results demonstrated that our novel ensemble approach achieved high accuracy when compared to existing work.

Predictive models for delay in medical decision-making among older patients with acute ischemic stroke: a comparative study using logistic regression analysis and lightGBM algorithm.

To explore the factors affecting delayed medical decision-making in older patients with acute ischemic stroke (AIS) using logistic regression analysis and the Light Gradient Boosting Machine (LightGBM) algorithm, and compare the two predictive models. A cross-sectional study was conducted among 309 older patients aged ≥ 60 who underwent AIS. Demographic characteristics, stroke onset characteristics, previous stroke knowledge level, health literacy, and social network were recorded. These data were separately inputted into logistic regression analysis and the LightGBM algorithm to build the predictive models for delay in medical decision-making among older patients with AIS. Five parameters of Accuracy, Recall, F1 Score, AUC and Precision were compared between the two models. The medical decision-making delay rate in older patients with AIS was 74.76%. The factors affecting medical decision-making delay, identified through logistic regression and LightGBM algorithm, were as follows: stroke severity, stroke recognition, previous stroke knowledge, health literacy, social network (common factors), mode of onset (logistic regression model only), and reaction from others (LightGBM algorithm only). The LightGBM model demonstrated the more superior performance, achieving the higher AUC of 0.909. This study used advanced LightGBM algorithm to enable early identification of delay in medical decision-making groups in the older patients with AIS. The identified influencing factors can provide critical insights for the development of early prevention and intervention strategies to reduce delay in medical decisions-making among older patients with AIS and promote patients' health. The LightGBM algorithm is the optimal model for predicting the delay in medical decision-making among older patients with AIS.

Comparison of Logistic Regression, Random Forest and Adaboost Algorithms for Diabetes Mellitus Classification

Diabetes mellitus is a chronic disease that affects the way the body regulates sugar (glucose). High blood sugar levels can lead to health complications including heart problems, eye disorders, nerve damage, kidney and blood vessel disorders. It is important for early detection of diabetes by utilizing data mining technology. Data mining has various classification models that can be used to detect diabetes, including logistic regression, random forest and adaboost. The comparison of the three algorithms aims to find out which algorithm is most appropriate in the classification of diabetes. From the results obtained, the random forest algorithm has the best performance in the classification of diabetes mellitus compared to other algorithms.

Intelligent Intrusion Detection System in Internet of Vehicles using Random Forest, Logistic Regression and Decision Tree Algorithm

This study introduces an Intelligent Intrusion Detection System (IDS) for Internet of Vehicles (IoV) security. Using Random Forest, Logistic Regression, and Decision Tree algorithms, it detects and prevents intrusions in real-time. By combining these algorithms, the IDS aims to improve accuracy and adapt to evolving threats. Evaluation metrics include accuracy (93.6%), precision (94.4%), recall (95.3%), and F1 score (96.7%). Analysis of false positives/negatives informs refinement for real-world use. The IDS enhances cybersecurity for connected vehicles. This IDS provides a robust defense against cyber threats in the IoV ecosystem, ensuring user safety and privacy. Its multi-algorithmic approach and high-performance metrics make it a reliable solution for safeguarding connected vehicles.

#2067 Prediction of gastrointestinal bleeding hospitalization risk in hemodialysis using machine learning techniques

Abstract Background and Aims INitiativeS on advancing Patients’ outcomes In REnal disease (INSPIRE) is an academia and industry collaboration set forth to identify critical investigations needed to advance the practice of medicine in nephrology. Gastrointestinal bleeding (GIB) is one of the most common types of bleeding events in the kidney dialysis population [1]. We aimed to develop a model to predict GIB hospitalization risk in a hemodialysis (HD) patient within 180 days. We evaluated advanced machine learning algorithm (XGBoost) and compared it to traditional machine learning algorithm (logistic regression) modeling techniques. Method We used data from a kidney care network from 2017 through 2020. We included data from adult dialysis patients (age ≥18 years). GIB related hospitalization was defined based on international classification of diseases (ICD) diagnosis codes recorded as the primary, secondary, or tertiary discharge reason for hospitalization. Two distinct models were created using XGBoost and logistic regression algorithms and using the same dataset. Both models were evaluated using metrics such as area under the receiver operating curve (AUROC), accuracy, sensitivity, and specificity. Missing data was imputed using mean for quantitative data and mode for qualitative data. The dataset then was randomly divided into 60% training, 20% validation and 20% test dataset. The test data, comprising unseen patients and data was used to evaluate the performance of the model. This means that the model had never encountered the test data during its learning phase, which included training the model and then validating the model. Results The incidence of 180-day GIB hospitalization was 1.12% in the HD population (n = 5 116 with GIB hospitalization/ n = 451 653 without GIB hospitalization), and consistent in the unseen test dataset (n = 465 with GIB hospitalization/ n = 38586 without GIB hospitalization). The XGBoost model showed higher predictive performance compared to logistic regression (Table). The AUROC was 0.72 (95% confidence interval (CI) 0.69, 0.74) for the XGBoost model. In comparison, the AUROC was 0.615 (95% CI 0.60, 0.64) for the logistic regression model. Specificity was 67% versus 58% for the XGBoost model versus logistic regression model respectively. However, sensitivity for both the models was 65%. The top predictors for major GIB were consistent in both models for many factors, yet some factors showed different importance on outcome prediction (Fig. 1). Higher risk of GIB hospitalization was associated with older age, lower ferritin levels, and recent all-cause hospitalizations in both models. The XGBoost showed high importance on outcome prediction for lower hemoglobin and higher serum 25 hydroxy (25OH) vitamin D values, and the logistic regression model showed high importance for higher amounts of saline delivered during an HD treatment and lower intact parathyroid hormone (iPTH) levels. Conclusion We found advanced machine learning prediction modeling (XGBoost) appears suitable for identifying a HD patient at risk for a GIB hospitalization in the next 180 days, and outperforms traditional machine learning (logistic regression) modeling techniques. Although both models showed identical sensitivity, the XGBoost model had higher specificity. Prospective testing is needed to confirm the model's performance. The association between bone mineral metabolism markers and GIB hospitalization risk is unexpected and warrants further investigation.

Forecasting Dairy Sales Estimates: A Comparative Analysis of Logistic Regression and Random Forest Algorithm

This study examines the effectiveness of logistic regression and random forest algorithms in forecasting sales estimates for dairy products. Using a comprehensive data set that includes factors such as product features, pricing dynamics, promotional efforts and consumer demographics, the models are trained to accurately predict future sales figures. Logistic regression provides a transparent framework for estimating probabilities, while random forest uses ensemble learning to capture complex relationships between variables. Through careful evaluation and comparison, the research aims to identify the strengths and weaknesses of each algorithm in producing reliable sales forecasts for dairy products. The results show that while logistic regression offers interpretability and simplicity, random forest excels in handling non-linear relationships and achieving higher prediction accuracy. Insights gathered from this analysis can help dairy industry stakeholders make informed decisions, optimize resource allocation and improve sales forecasting strategies.

Privacy-Preserving Breast Cancer Prediction Based on Logistic Regression

Abstract With the increasing strain on today’s healthcare resources, there is a growing demand for pre-diagnosis testing. In response, researchers have suggested diverse machine learning models for disease prediction, among which logistic regression stands out as one of the most effective models. Its objective is to enhance the accuracy and efficiency of pre-diagnosis testing, thereby alleviating the burden on healthcare resources. However, when multiple medical institutions collaborate to train models, the untrusted cloud server may pose a risk of private data leakage, enabling participants to steal data from one another. Existing privacy-preserving methods often suffer from drawbacks such as high communication costs, long training times and lack of security proofs. Therefore, it is imperative to jointly train an excellent model collaboratively and uphold data privacy. In this paper, we develop a highly optimized two-party logistic regression algorithm based on CKKS scheme. The algorithm optimizes ciphertext operations by employing ciphertext segmentation and minimizing the multiplication depth, resulting in time savings. Furthermore, it utilizes least squares to approximate sigmoid functions within specific intervals that cannot be handled by homomorphic encryption. Finally, the proposed algorithm is evaluated on a breast cancer dataset, and simulation experiments demonstrate that the model’s prediction accuracy, after machine learning training, exceeds 96% for two-sided encrypted data.

Integration of Multiple Spectral Data via a Logistic Regression Algorithm for Detection of Crop Residue Burned Areas: A Case Study of Songnen Plain, Northeast China

Integration of Multiple Spectral Data via a Logistic Regression Algorithm for Detection of Crop Residue Burned Areas: A Case Study of Songnen Plain, Northeast China

SPOTTING OF CYBERBULLYING ON SOCIAL MEDIA USING MACHINE LEARNING

Cyberbullying occurs when someone harasses, threatens, or mistreats someone else online or on social media. Cyberbullying leads to threats, public embarrassment, and reprimands. Cyberbullying has caused a rise in youth suicide and mental health issues. It has lowered my self-esteem and escalated my suicidal thoughts. If cyberbullying continues, an entire generation of young adults will suffer from low self-esteem and mental health issues. Many machine learning algorithms are used to automatically identify social media cyberbullying. Social media monitoring allows this. These models don't account for all the factors that could make a comment or post bullying. Bullying is still ambiguous. This study presents a cyberbullying diagnosis model based on several factors. Cyberbully is a misuse of technology advantage to bully a person. This project is used Dataset namely,’ Cyberbullying-tweets’. This project used Machine Learning Classifiers such as Support Vector Machine (SVM), Random Forest (RF), Naïve Bayes (NB), and Neural Networks (NN), Logistic Regression(LR) Algorithms and calculated performance results for comparing the performance for dataset. Keywords: Cyberbullying Detection, Social Media Analysis, Machine Learning Models, Natural Language Processing (NLP), Text Classification , Feature Extraction, Sentiment Analysis, Textual Data Mining, Online Harassment, User Profiling, Toxic Language Detection, Hate Speech Detection, Content Moderation, Multi-Layer Perceptron (MLP), Support Vector Machine (SVM), Neural Network Algorithms, Lexicon-based Features, Syntactic Features, N-gram Analysis, Arabic Social Media (if focusing on Arabic language),User Behavior Analysis, Feature Engineering, Cross- Platform Analysis, Real-time Monitoring, Social Network Analysis

Machine learning models to predict submucosal invasion in early gastric cancer based on endoscopy features and standardized color metrics

Conventional endoscopy is widely used in the diagnosis of early gastric cancers (EGCs), but the graphical features were loosely defined and dependent on endoscopists’ experience. We aim to establish a more accurate predictive model for infiltration depth of early gastric cancer including a standardized colorimetric system, which demonstrates promising clinical implication. A retrospective study of 718 EGC cases was performed. Clinical and pathological characteristics were included, and Commission Internationale de l’Eclariage (CIE) standard colorimetric system was used to evaluate the chromaticity of lesions. The predicting models were established in the derivation set using multivariate backward stepwise logistic regression, decision tree model, and random forest model. Logistic regression shows location, macroscopic type, length, marked margin elevation, WLI color difference and histological type are factors significantly independently associated with infiltration depth. In the decision tree model, margin elevation, lesion located in the lower 1/3 part, WLI a*color value, b*color value, and abnormal thickness in enhanced CT were selected, which achieved an AUROC of 0.810. A random forest model was established presenting the importance of each feature with an accuracy of 0.80, and an AUROC of 0.844. Quantified color metrics can improve the diagnostic precision in the invasion depth of EGC. We have developed a nomogram model using logistic regression and machine learning algorithms were also explored, which turned out to be helpful in decision-making progress.

Twitter sentiment analysis for Hausa abbreviations and acronyms

The use of natural language processing, to identify, extract and organize sentiment from user generated texts in social networks, blogs or product review of text is known as sentiment analysis or opinion mining. Hausa language belongs to one of the major well-spoken languages in Africa and one of the three major Nigerian languages. Now investigating into such a language will have significant influence on social, economic business political and even educational services and settings. Some of these Hausa texts are abbreviated and some in acronym format which is a challenge to researchers as such comments are in an unstructured format and needs normalization to get further understanding of that text and also there is scarcity of sentiment analysis on Hausa abbreviation and acronym. Abbreviation is a shorten form of a word while acronym is an abbreviation formed from the initial letters of other words and pronounced as a word. This research aims to develop an improved Hausa Sentiment Dataset for the enhancement of sentiment analysis with abbreviation and acronyms. This is achieved by adapting to the approach for Hausa Sentiment Analysis based on Multinomial Naïve Bayes (MNB) and Logistic Regression algorithms using the count vectorizer, along with python libraries for NLP. This research affirmed that the improved dataset with abbreviation and acronym outperforms the plain Hausa dataset by 4% in accuracy using Multinomial Naïve Bayes. The result shows that in addition to normal preprocessing techniques of the social media stream, understanding, interpreting and resolving ambiguity in the usage of abbreviations and acronyms lead to improved accuracy of algorithms with evidence in the experimental result.

Exploration of the application potential of serum multi-biomarker model in colorectal cancer screening

Analyzing blood lipid and bile acid profile changes in colorectal cancer (CRC) patients. Evaluating the integrated model's diagnostic significance for CRC. Ninety-one individuals with colorectal cancer (CRC group) and 120 healthy volunteers (HC group) were selected for comparison. Serum levels of total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), and apolipoproteins (Apo) A1, ApoA2, ApoB, ApoC2, and ApoC3 were measured using immunoturbidimetric and colorimetric methods. Additionally, LC–MS/MS was employed to detect fifteen bile acids in the serum, along with six tumor markers: carcinoembryonic antigen (CEA), carbohydrate antigens (CA) 125, CA19-9, CA242, CA50, and CA72-4. Group comparisons utilized independent sample t-tests and Mann–Whitney U tests. A binary logistic regression algorithm was applied to fit the indicators and establish a screening model; the diagnostic accuracy of individual Indicators and the model was analyzed using receiver operating characteristic (ROC) curves. The CRC group showed significantly lower levels in eight serum lipid indicators and eleven bile acids compared to the HC group (P < 0.05). Conversely, serum levels of TG, CA19-9, and CEA were elevated (P < 0.05). Among the measured parameters, ApoA2 stands out for its strong correlation with the presence of CRC, showcasing exceptional screening efficacy with an area under the curve (AUC) of 0.957, a sensitivity of 85.71%, and a specificity of 93.33%. The screening model, integrating ApoA1, ApoA2, lithocholic acid (LCA), and CEA, attained an impressive AUC of 0.995, surpassing the diagnostic accuracy of individual lipids, bile acids, and tumor markers. CRC patients manifest noteworthy alterations in both blood lipids and bile acid profiles. A screening model incorporating ApoA1, ApoA2, LCA, and CEA provides valuable insights for detecting CRC.

Chest Computed Tomography Radiomics for Determining Macrolide Resistance-Associated Gene Mutation Status in Children with Mycoplasma pneumoniae Pneumonia: A Two-Center Study

Rationale and ObjectivesThe mutations in the 23S ribosomal RNA (rRNA) gene are associated with an increase in resistance to macrolides in children with Mycoplasma pneumoniae pneumonia (MPP). This study aimed to develop and validate a chest computed tomography (CT) radiomics model for determining macrolide resistance-associated gene mutation status in MPP. Materials and MethodsA total of 258 MPP patients were retrospectively included from two institutions (training set: 194 patients from the first institution; external test set: 64 patients from the second). The 23S rRNA gene mutation status was tested by nasopharyngeal swab polymerase chain reaction. Radiomics features were extracted from chest CT images of pulmonary lesions segmented with semi-automatic delineation. Subsequently, radiomics feature reduction was applied to identify the most relevant features. Logistic regression and random forest algorithms were employed to establish the radiomics models, which were five-fold cross-validated in the training set and validated in the external test set. ResultsThe radiomics feature selection resulted in eight features. After five-fold cross-validation in the training set, the mean areas under the receiver operating characteristic curve (AUCs) of the logistic regression and random forest models were 0.868 (95% confidence interval (CI): 0.813-0.923) and 0.941 (95% CI: 0.907-0.975), respectively. In the external test set, the corresponding AUCs were 0.855 (95% CI: 0.758-0.952) and 0.815 (95% CI: 0.705-0.925). ConclusionChest CT radiomics is a promising diagnostic tool for determining macrolide resistance gene mutation status in MPP. Availability of data and materialThe datasets generated or analyzed during the study are available from the corresponding author on reasonable request.

Design of antiviral AGO2-dependent short hairpin RNAs

The increasing emergence and re-emergence of RNA virus outbreaks underlines the urgent need to develop effective antivirals. RNA interference (RNAi) is a sequence-specific gene silencing mechanism that is triggered by small interfering RNAs (siRNAs) or short hairpin RNAs (shRNAs), which exhibits significant promise for antiviral therapy. AGO2-dependent shRNA (agshRNA) generates a single-stranded guide RNA and presents significant advantages over traditional siRNA and shRNA. In this study, we applied a logistic regression algorithm to a previously published chemically siRNA efficacy dataset and built a machine learning-based model with high predictive power. Using this model, we designed siRNA sequences targeting diverse RNA viruses, including human enterovirus A71 (EV71), Zika virus (ZIKV), dengue virus 2 (DENV2), mouse hepatitis virus (MHV) and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and transformed them into agshRNAs. We validated the performance of our agshRNA design by evaluating antiviral efficacies of agshRNAs in cells infected with different viruses. Using the agshRNA targeting EV71 as an example, we showed that the anti-EV71 effect of agshRNA was more potent compared with the corresponding siRNA and shRNA. Moreover, the antiviral effect of agshRNA is dependent on AGO2-processed guide RNA, which can load into the RNA-induced silencing complex (RISC). We also confirmed the antiviral effect of agshRNA in vivo. Together, this work develops a novel antiviral strategy that combines machine learning-based algorithm with agshRNA design to custom design antiviral agshRNAs with high efficiency.

Personalized Recommendation of English Online Teaching Content Based on Logistic Regression Algorithm

English online teaching content encompasses a wide range of materials designed to facilitate language learning in digital environments. These resources include interactive lessons, video tutorials, audio recordings, quizzes, vocabulary exercises, grammar explanations, and authentic texts such as articles, short stories, and dialogues. The content aims to cover various language skills, including listening, speaking, reading, and writing, catering to learners of different proficiency levels and learning styles. This paper introduces a Cross-Domain Personalized Recommendation System (CDPRS) designed to enhance the delivery of English language teaching content. Leveraging machine learning algorithms and data analytics techniques, the CDPRS tailors recommendations to individual user preferences, learning objectives, and proficiency levels across diverse content domains. Through a series of analyses and evaluations, including regression analysis and simulation studies, the effectiveness and utility of the CDPRS are demonstrated. Through a series of analyses and evaluations, including regression analysis and simulation studies, the effectiveness and utility of the CDPRS are demonstrated. Results indicate that the system achieves high levels of accuracy (mean accuracy = 0.85), diversity (mean diversity = 0.72), novelty (mean novelty = 0.78), and user satisfaction (mean satisfaction = 0.82) in providing personalized recommendations.

Workload prediction based on improved error correlation logistic regression algorithm and Cross‐TRCN of spatiotemporal neural network

AbstractIn view of the randomness of user network usage behavior in data centers, which leads to a large randomness in power load, and considering that a single randomness processing method is usually difficult to fully characterize the uncertain characteristics of the system, this paper proposes a dual fusion prediction analysis model based on an improved error correlation logic regression algorithm and a novel spatiotemporal neural network structure called Cross‐TRCN. Two weight coefficients λ1 and λ2 are introduced to fuse the prediction results with different long‐term sequence prediction performance, thereby further eliminating the influence of random errors. The results show that it is feasible to predict the workload of data centers based on the improved error correlation logic regression algorithm and the innovative spatiotemporal neural network structure Cross‐TRCN.

To identify important MRI features to differentiate hepatic mucinous cystic neoplasms from septated hepatic cysts based on random forest.

To identify important MRI features to differentiate hepatic mucinous cystic neoplasms (MCN) from septated hepatic cysts (HC) using random forest andcompared with logistic regression algorithm. Pathologically diagnosed hepatic cysts and hepatic MCNs with pre-operative contrast-enhanced MRI in our hospital from 2010 to 2023 were collected and only septated lesions on enhanced MRI were enrolled. A total of 21 septated HC and 18 MCNs were included in this study. Eighteen MRI features were analyzed and top important features were identified based on random forest (RF) algorithm. The results were evaluated by the prediction performance of a RF model combining the important features and compared with the performance of the logistic regression (LR) algorithm. Finally, for each identified feature, diagnostic probability, sensitivity, and specificity were calculated and compared. Four variables, i.e., the septation arising from wall without indentation, multiseptate, intracapsular cyst sign, and solitary lesion were extracted as top important features with significance for MCNs by the random forest algorithm. The RF model using these variables had an AUC of 0.982 (0.95CI, 0.950-1.000), compared with the LR model based on two identified features with AUC of 0.931 (0.95CI, 0.846-1.000), p = 0.202. Among the four important features, multiseptate had the highest specificity (95.2%) and good sensitivity (72.2%, lower than the septation from wall without indentation, 94.4%) to diagnose MCNs. Four out of 18 MRI features were extracted as reliably important factors to differ hepatic MCNs from septated HC. The combination of these four features in a RF model could achieve satisfactory diagnostic efficacy.

Web Attack Detection Based on Honeypots and Logistic Regression Algorithm

Web security has emerged as one of the most prominent concerns in the realm of cybersecurity. Traditional rule-based methods for detecting web attacks often rely on manual rule definition and pattern matching, leaving them inadequate in accurately identifying new and intricate attack patterns. In the face of these challenges, machine learning techniques have demonstrated potential and advantages. This paper presents a web attack detection method based on honeypots and the logistic regression algorithm. It involves the cleansing, filtering, and analysis of web logs captured by honeypots, followed by the vectorization of the textual data contained in these logs. The logistic regression algorithm is employed to train and test the classification of the text vectors, generating a logistic regression model. This model is then used to predict newly generated web logs, enabling effective dynamic web attack detection. Experimental evaluations using collected datasets are conducted, comparing the proposed method with the support vector machine approach. The results demonstrate that this method achieves rapid and accurate detection and recognition of web attack behaviors while ensuring performance efficacy.

SPAMMER DETECTION AND FAKE USER IDENTIFICATION ON SOCIAL NETWORKS

The project titled "Spammer Detection and Fake User Identification on Social Networks" aims to explore and implement a novel approach to concealing information within digital images while preserving their visual integrity. With the exponential growth of social media platforms like Instagram, the issue of spam and fake user accounts has become increasingly prevalent, posing significant challenges to the integrity and user experience of these platforms. In this project, a comprehensive approach to address the problem of spammer detection and fake user identification on Instagram is proposed. Leveraging machine learning techniques, including Decision Tree Classifier, KNN Classifier, Random Forest Classifier, and Logistic Regression Algorithm, the project aims to develop robust models capable of automatically identifying and flagging suspicious accounts. By analyzing various features such as user behaviour patterns, engagement metrics, and content characteristics, these classifiers will be trained to differentiate between genuine and fake accounts effectively. The project's ultimate goal is to contribute to the enhancement of Instagram's spam detection mechanisms, fostering a safer and more authentic social media environment for users. Key Words: Spam detection, Fake user identification, KNN Classifier, Decision Tree Classifier, Random Forest Classifier, Logistic Regression Algorithm.

Combined blood Neurofilament light chain and third ventricle width to differentiate Progressive Supranuclear Palsy from Parkinson's Disease: A machine learning study

IntroductionDifferentiating Progressive Supranuclear Palsy (PSP) from Parkinson's Disease (PD) may be clinically challenging. In this study, we explored the performance of machine learning models based on MR imaging and blood molecular biomarkers in distinguishing between these two neurodegenerative diseases. MethodsTwenty-eight PSP patients, 46 PD patients and 60 control subjects (HC) were consecutively enrolled in the study. Serum concentration of neurofilament light chain protein (Nf-L) was assessed by single molecule array (SIMOA), while an automatic segmentation algorithm was employed for T1-weighted measurements of third ventricle width/intracranial diameter ratio (3rdV/ID). Machine learning (ML) models with Logistic Regression (LR), Random Forest (RF), and XGBoost algorithms based on 3rdV/ID and serum Nf-L levels were tested in distinguishing among PSP, PD and HC. ResultsPSP patients showed higher serum Nf-L levels and larger 3rdV/ID ratio in comparison with both PD and HC groups (p < 0.005). All ML algorithms (LR, RF and XGBoost) showed that the combination of MRI and blood biomarkers had excellent classification performances in differentiating PSP from PD (AUC ≥0.92), outperforming each biomarker used alone (AUC: 0.85–0.90). Among the different algorithms, XGBoost was slightly more powerful than LR and RF in distinguishing PSP from PD patients, reaching AUC of 0.94 ± 0.04. ConclusionOur findings highlight the usefulness of combining blood and simple linear MRI biomarkers to accurately distinguish between PSP and PD patients. This multimodal approach may play a pivotal role in patient management and clinical decision-making, paving the way for more effective and timely interventions in these neurodegenerative diseases.