BioMedical Engineering OnLine最新文献

{"title":"Examining temporal changes in model-optimized parameters using longitudinal hemodynamic measurements.","authors":"Nikolai L Bjørdalsbakke, Jacob Sturdy, Ulrik Wisløff, Leif R Hellevik","doi":"10.1186/s12938-024-01242-y","DOIUrl":"10.1186/s12938-024-01242-y","url":null,"abstract":"<p><strong>Background: </strong>We previously applied hemodynamic data to personalize a mathematical model of the circulation expressed as physically interpretable parameters. The aim of this study was to identify patterns in the data that could potentially explain the estimated parameter changes. This included investigating whether the parameters could be used to track the effect of physical activity on high blood pressure. Clinical trials have repeatedly detected beneficial changes in blood pressure after physical activity and uncovered changes in lower level phenotypes (such as stiffened or high-resistance blood vessels). These phenotypes can be characterized by parameters describing the mechanical properties of the circulatory system. These parameters can be incorporated in and contextualized by physics-based cardiovascular models of the circulation, which in combination can become tools for monitoring cardiovascular disease progression and management in the future.</p><p><strong>Methods: </strong>Closed-loop and open-loop models of the left ventricle and systemic circulation were previously optimized to data from a pilot study with a 12-week exercise intervention period. Basal characteristics and hemodynamic data such as blood pressure in the carotid, brachial and finger arteries, as well as left-ventricular outflow tract flow traces were collected in the trial. Model parameters estimated for measurements made on separate days during the trial were used to compute parameter changes for total peripheral resistance, systemic arterial compliance, and maximal left-ventricular elastance. We compared the changes in these cardiovascular model-based estimates to changes from more conventional estimates made without the use of physics-based models by correlation analysis. Additionally, ordinary linear regression and linear mixed-effects models were applied to determine the most informative measurements for the selected parameters. We applied maximal aerobic capacity (measured as <math><mtext>VO2max</mtext></math> ) data to examine if exercise had any impact on parameters through regression analysis and case studies.</p><p><strong>Results and conclusions: </strong>Parameter changes in arterial parameters estimated using the cardiovascular models correlated moderately well with conventional estimates. Estimates based on carotid pressure waveforms gave higher correlations (0.59 and above when p <math><mrow><mo><</mo> <mn>0.05</mn></mrow> </math> ) than those for finger arterial pressure. Parameter changes over the 12-week study duration were of similar magnitude when compared to short-term changes after a bout of intensive exercise in the same parameters. The short-term changes were computed from measurements made immediately before and 24 h after a cardiopulmonary exercise test used to measure <math><mtext>VO2max</mtext></math> . Regression analysis indicated that changes in <math><mtext>VO2max</mtext></math> did not account for any substantial amount ","PeriodicalId":8927,"journal":{"name":"BioMedical Engineering OnLine","volume":"23 1","pages":"64"},"PeriodicalIF":2.9,"publicationDate":"2024-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11234604/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141562537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automatic localization of cochlear implant electrodes using cone beam computed tomography images","authors":"Jasmin Thormählen, Benjamin Krüger, Waldo Nogueira","doi":"10.1186/s12938-024-01249-5","DOIUrl":"https://doi.org/10.1186/s12938-024-01249-5","url":null,"abstract":"Cochlear implants (CI) are implantable medical devices that enable the perception of sounds and the understanding of speech by electrically stimulating the auditory nerve in case of inner ear damage. The stimulation takes place via an array of electrodes surgically inserted in the cochlea. After CI implantation, cone beam computed tomography (CBCT) is used to evaluate the position of the electrodes. Moreover, CBCT is used in research studies to investigate the relationship between the position of the electrodes and the hearing outcome of CI user. In clinical routine, the estimation of the position of the CI electrodes is done manually, which is very time-consuming. The aim of this study was to optimize procedures of automatic electrode localization from CBCT data following CI implantation. For this, we analyzed the performance of automatic electrode localization for 150 CBCT data sets of 10 different types of electrode arrays. Our own implementation of the method by Noble and Dawant (Lecture notes in computer science (Including subseries lecture notes in artificial intelligence and lecture notes in bioinformatics), Springer, pp 152–159, 2015. https://doi.org/10.1007/978-3-319-24571-3_19 ) for automated electrode localization served as a benchmark for evaluation. Differences in the detection rate and the localization accuracy across types of electrode arrays were evaluated and errors were classified. Based on this analysis, we developed a strategy to optimize procedures of automatic electrode localization. It was shown that particularly distantly spaced electrodes in combination with a deep insertion can lead to apical–basal confusions in the localization procedure. This confusion prevents electrodes from being detected or assigned correctly, leading to a deterioration in localization accuracy. We propose an extended cost function for automatic electrode localization methods that prevents double detection of electrodes to avoid apical–basal confusions. This significantly increased the detection rate by 11.15 percent points and improved the overall localization accuracy by 0.53 mm (1.75 voxels). In comparison to other methods, our proposed cost function does not require any prior knowledge about the individual cochlea anatomy.","PeriodicalId":8927,"journal":{"name":"BioMedical Engineering OnLine","volume":"31 1","pages":""},"PeriodicalIF":3.9,"publicationDate":"2024-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141577392","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Measures of overnight oxygen saturation to characterize sleep apnea severity and predict postoperative respiratory depression.","authors":"Atousa Assadi, Frances Chung, Azadeh Yadollahi","doi":"10.1186/s12938-024-01254-8","DOIUrl":"10.1186/s12938-024-01254-8","url":null,"abstract":"<p><strong>Background: </strong>Sleep apnea syndrome, characterized by recurrent cessation (apnea) or reduction (hypopnea) of breathing during sleep, is a major risk factor for postoperative respiratory depression. Challenges in sleep apnea assessment have led to the proposal of alternative metrics derived from oxyhemoglobin saturation (SpO₂), such as oxygen desaturation index (ODI) and percentage of cumulative sleep time spent with SpO₂ below 90% (CT90), as predictors of postoperative respiratory depression. However, their performance has been limited with area under the curve of 0.60 for ODI and 0.59 for CT90. Our objective was to propose novel features from preoperative overnight SpO₂ which are correlated with sleep apnea severity and predictive of postoperative respiratory depression.</p><p><strong>Methods: </strong>Preoperative SpO₂ signals from 235 surgical patients were retrospectively analyzed to derive seven features to characterize the sleep apnea severity. The features included entropy and standard deviation of SpO₂ signal; below average burden characterizing the area under the average SpO₂; average, standard deviation, and entropy of desaturation burdens; and overall nocturnal desaturation burden. The association between the extracted features and sleep apnea severity was assessed using Pearson correlation analysis. Logistic regression was employed to evaluate the predictive performance of the features in identifying postoperative respiratory depression.</p><p><strong>Results: </strong>Our findings indicated a similar performance of the proposed features to the conventional apnea-hypopnea index (AHI) for assessing sleep apnea severity, with average area under the curve ranging from 0.77 to 0.81. Notably, entropy and standard deviation of overnight SpO₂ signal and below average burden showed comparable predictive capability to AHI but with minimal computational requirements and individuals' burden, making them promising for screening purposes. Our sex-based analysis revealed that compared to entropy and standard deviation, below average burden exhibited higher sensitivity in detecting respiratory depression in women than men.</p><p><strong>Conclusion: </strong>This study underscores the potential of preoperative SpO₂ features as alternative metrics to AHI in predicting postoperative respiratory.</p>","PeriodicalId":8927,"journal":{"name":"BioMedical Engineering OnLine","volume":"23 1","pages":"63"},"PeriodicalIF":2.9,"publicationDate":"2024-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11229251/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141557968","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Nuclear proteins and diabetic retinopathy: a review.","authors":"Bin Li, Wahab Hussain, Zhi-Liang Jiang, Jia-Yi Wang, Sarfraz Hussain, Talat Bilal Yasoob, Yuan-Kun Zhai, Xin-Ying Ji, Ya-Long Dang","doi":"10.1186/s12938-024-01258-4","DOIUrl":"10.1186/s12938-024-01258-4","url":null,"abstract":"<p><p>Diabetic retinopathy (DR) is an eye disease that causes blindness and vision loss in diabetic. Risk factors for DR include high blood glucose levels and some environmental factors. The pathogenesis is based on inflammation caused by interferon and other nuclear proteins. This review article provides an overview of DR and discusses the role of nuclear proteins in the pathogenesis of the disease. Some core proteins such as MAPK, transcription co-factors, transcription co-activators, and others are part of this review. In addition, some current advanced treatment resulting from the role of nuclear proteins will be analyzes, including epigenetic modifications, the use of methylation, acetylation, and histone modifications. Stem cell technology and the use of nanobiotechnology are proposed as promising approaches for a more effective treatment of DR.</p>","PeriodicalId":8927,"journal":{"name":"BioMedical Engineering OnLine","volume":"23 1","pages":"62"},"PeriodicalIF":2.9,"publicationDate":"2024-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11197269/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141449622","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Peak detection in intracranial pressure signal waveforms: a comparative study.","authors":"Miaomiao Wei, Solventa Krakauskaite, Sreya Subramanian, Fabien Scalzo","doi":"10.1186/s12938-024-01245-9","DOIUrl":"10.1186/s12938-024-01245-9","url":null,"abstract":"<p><strong>Background: </strong>The monitoring and analysis of quasi-periodic biological signals such as electrocardiography (ECG), intracranial pressure (ICP), and cerebral blood flow velocity (CBFV) waveforms plays an important role in the early detection of adverse patient events and contributes to improved care management in the intensive care unit (ICU). This work quantitatively evaluates existing computational frameworks for automatically extracting peaks within ICP waveforms.</p><p><strong>Methods: </strong>Peak detection techniques based on state-of-the-art machine learning models were evaluated in terms of robustness to varying noise levels. The evaluation was performed on a dataset of ICP signals assembled from 700 h of monitoring from 64 neurosurgical patients. The groundtruth of the peak locations was established manually on a subset of 13, 611 pulses. Additional evaluation was performed using a simulated dataset of ICP with controlled temporal dynamics and noise.</p><p><strong>Results: </strong>The quantitative analysis of peak detection algorithms applied to individual waveforms indicates that most techniques provide acceptable accuracy with a mean absolute error (MAE) <math><mrow><mo>≤</mo> <mn>10</mn></mrow> </math> ms without noise. In the presence of a higher noise level, however, only kernel spectral regression and random forest remain below that error threshold while the performance of other techniques deteriorates. Our experiments also demonstrated that tracking methods such as Bayesian inference and long short-term memory (LSTM) can be applied continuously and provide additional robustness in situations where single pulse analysis methods fail, such as missing data.</p><p><strong>Conclusion: </strong>While machine learning-based peak detection methods require manually labeled data for training, these models outperform conventional signal processing ones based on handcrafted rules and should be considered for peak detection in modern frameworks. In particular, peak tracking methods that incorporate temporal information between successive periods of the signals have demonstrated in our experiments to provide more robustness to noise and temporary artifacts that commonly arise as part of the monitoring setup in the clinical setting.</p>","PeriodicalId":8927,"journal":{"name":"BioMedical Engineering OnLine","volume":"23 1","pages":"61"},"PeriodicalIF":2.9,"publicationDate":"2024-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11194974/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141445315","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Pulse wave signal-driven machine learning for identifying left ventricular enlargement in heart failure patients.","authors":"Dandan Wu, Ryohei Ono, Sirui Wang, Yoshio Kobayashi, Koichi Sughimoto, Hao Liu","doi":"10.1186/s12938-024-01257-5","DOIUrl":"10.1186/s12938-024-01257-5","url":null,"abstract":"<p><strong>Background: </strong>Left ventricular enlargement (LVE) is a common manifestation of cardiac remodeling that is closely associated with cardiac dysfunction, heart failure (HF), and arrhythmias. This study aimed to propose a machine learning (ML)-based strategy to identify LVE in HF patients by means of pulse wave signals.</p><p><strong>Method: </strong>We constructed two high-quality pulse wave datasets comprising a non-LVE group and an LVE group based on the 264 HF patients. Fourier series calculations were employed to determine if significant frequency differences existed between the two datasets, thereby ensuring their validity. Then, the ML-based identification was undertaken by means of classification and regression models: a weighted random forest model was employed for binary classification of the datasets, and a densely connected convolutional network was utilized to directly estimate the left ventricular diastolic diameter index (LVDdI) through regression. Finally, the accuracy of the two models was validated by comparing their results with clinical measurements, using accuracy and the area under the receiver operating characteristic curve (AUC-ROC) to assess their capability for identifying LVE patients.</p><p><strong>Results: </strong>The classification model exhibited superior performance with an accuracy of 0.91 and an AUC-ROC of 0.93. The regression model achieved an accuracy of 0.88 and an AUC-ROC of 0.89, indicating that both models can quickly and accurately identify LVE in HF patients.</p><p><strong>Conclusion: </strong>The proposed ML methods are verified to achieve effective classification and regression with good performance for identifying LVE in HF patients based on pulse wave signals. This study thus demonstrates the feasibility and potential of the ML-based strategy for clinical practice while offering an effective and robust tool for diagnosing and intervening ventricular remodeling.</p>","PeriodicalId":8927,"journal":{"name":"BioMedical Engineering OnLine","volume":"23 1","pages":"60"},"PeriodicalIF":2.9,"publicationDate":"2024-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11193305/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141440106","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Changes in temporal lobe activation during a sound stimulation task in patients with sensorineural tinnitus: a multi-channel near-infrared spectroscopy study.","authors":"Xiaoli Fan, Bin Gong, Hao Yang, Juanjuan Yang, Gaowei Qi, Zheng Wang, Jie Sun, Yu Fang","doi":"10.1186/s12938-024-01255-7","DOIUrl":"10.1186/s12938-024-01255-7","url":null,"abstract":"<p><strong>Background: </strong>The subjective sign of a serious pandemic in human work and life is mathematical neural tinnitus. fNIRS (functional near-infrared spectroscopy) is a new non-invasive brain imaging technology for studying the neurological activity of the human cerebral cortex. It is based on neural coupling effects. This research uses the fNIRS approach to detect differences in the neurological activity of the cerebral skin in the sound stimulation mission in order to better discriminate between the sensational neurological tinnitus.</p><p><strong>Methods: </strong>In the fNIRS brain imaging method, 14 sensorineural tinnitus sufferers and 14 healthy controls listened to varied noise and quiet for fNIRS data collection. Linear fitting was employed in MATLAB to eliminate slow drifts during preprocessing and event-related design analysis. The false discovery rate (FDR) procedure was applied in IBM SPSS Statistics 26.0 to control the false positive rate in multiple comparison analyses.</p><p><strong>Results: </strong>When the ill group and the healthy control group were stimulated by pink noise, there was a significant difference in blood oxygen concentration (P < 0.05), and the healthy control group exhibited a high activation, according to the fNIRS measurement data. The blood oxygen concentration level in the patient group was dramatically enhanced after one month of acupuncture therapy under the identical stimulation task settings, and it was favorably connected with the levels of THI and TEQ scales.</p><p><strong>Conclusions: </strong>Using sensorineural tinnitus illness as an example, fNIRS technology has the potential to disclose future pathological study on subjective diseases throughout time. Other clinical disorders involving the temporal lobe and adjacent brain areas may also be examined, in addition to tinnitus-related brain alterations.</p>","PeriodicalId":8927,"journal":{"name":"BioMedical Engineering OnLine","volume":"23 1","pages":"59"},"PeriodicalIF":2.9,"publicationDate":"2024-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11191237/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141431298","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"EfficientNet-based machine learning architecture for sleep apnea identification in clinical single-lead ECG signal data sets.","authors":"Meng-Hsuan Liu, Shang-Yu Chien, Ya-Lun Wu, Ting-Hsuan Sun, Chun-Sen Huang, Kai-Cheng Hsu, Liang-Wen Hang","doi":"10.1186/s12938-024-01252-w","DOIUrl":"10.1186/s12938-024-01252-w","url":null,"abstract":"<p><strong>Objective: </strong>Our objective was to create a machine learning architecture capable of identifying obstructive sleep apnea (OSA) patterns in single-lead electrocardiography (ECG) signals, exhibiting exceptional performance when utilized in clinical data sets.</p><p><strong>Methods: </strong>We conducted our research using a data set consisting of 1656 patients, representing a diverse demographic, from the sleep center of China Medical University Hospital. To detect apnea ECG segments and extract apnea features, we utilized the EfficientNet and some of its layers, respectively. Furthermore, we compared various training and data preprocessing techniques to enhance the model's prediction, such as setting class and sample weights or employing overlapping and regular slicing. Finally, we tested our approach against other literature on the Apnea-ECG database.</p><p><strong>Results: </strong>Our research found that the EfficientNet model achieved the best apnea segment detection using overlapping slicing and sample-weight settings, with an AUC of 0.917 and an accuracy of 0.855. For patient screening with AHI > 30, we combined the trained model with XGBoost, leading to an AUC of 0.975 and an accuracy of 0.928. Additional tests using PhysioNet data showed that our model is comparable in performance to existing models regarding its ability to screen OSA levels.</p><p><strong>Conclusions: </strong>Our suggested architecture, coupled with training and preprocessing techniques, showed admirable performance with a diverse demographic dataset, bringing us closer to practical implementation in OSA diagnosis. Trial registration The data for this study were collected retrospectively from the China Medical University Hospital in Taiwan with approval from the institutional review board CMUH109-REC3-018.</p>","PeriodicalId":8927,"journal":{"name":"BioMedical Engineering OnLine","volume":"23 1","pages":"57"},"PeriodicalIF":2.9,"publicationDate":"2024-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11188209/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141431300","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Diurnal variation of heart rate variability in individuals with spinal cord injury.","authors":"Jittima Saengsuwan, Arphatsorn Ruangsuphaphichat, Lars Brockmann, Patpiya Sirasaporn, Nuttaset Manimmanakorn, Kenneth J Hunt","doi":"10.1186/s12938-024-01256-6","DOIUrl":"10.1186/s12938-024-01256-6","url":null,"abstract":"<p><strong>Background: </strong>Heart rate variability (HRV) may provide objective information about cardiogenic autonomic balance in individuals with spinal cord injury (SCI). The aim of this study was to characterize the diurnal variation of HRV in individuals with SCI at lesion level T6 and above and lesion level below T6.</p><p><strong>Methods: </strong>This was a retrospective analysis of a prior cross-sectional study. Individuals with chronic SCI underwent 24 h recording of the time between consecutive R waves (RR interval) to derive parameters of HRV as follows: standard deviation of all normal-to-normal R-R intervals (SDNN) and square root of the mean of the squared differences between successive R-R intervals (RMSSD) (time domain); and high frequency power (HF), low-frequency power (LF), very low frequency power (VLF), ultra-low frequency power (ULF) and total power (TP) (frequency domain). Changes in the magnitude of HRV outcomes over the 24 h period were investigated using a novel multi-component cosinor model constrained to the form of a three-harmonic Fourier series.</p><p><strong>Results: </strong>Participants were grouped as lesion level T6 and above (n = 22) or below T6 (n = 36). Most of them were male (n = 40, 69%) and the median age (interquartile range) was 50.5 (28) years. Both groups exhibited similar diurnal patterns in most HRV metrics. The lowest values occurred in the late afternoon (4-6 pm) and gradually increased, peaking around midnight to early morning (1-6 am). Exceptions included RMSSD, which peaked before midnight, and ULF, which showed a double peak pattern that peaked from 11 am to 1 pm and 4-6 am in participants with lesion level at T6 and above. The HRV values in participants with lesion level T6 and above were generally lower than participants with lesion level below T6, except for peak values of RMSSD, HF and LF.</p><p><strong>Conclusion: </strong>This study demonstrated substantial diurnal variation of HRV in participants with SCI in both groups of participants. In clinical and research settings, diurnal variations in HRV must be taken into consideration.</p>","PeriodicalId":8927,"journal":{"name":"BioMedical Engineering OnLine","volume":"23 1","pages":"58"},"PeriodicalIF":2.9,"publicationDate":"2024-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11188279/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141431299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Noninvasive prediction of lymph node metastasis in pancreatic cancer using an ultrasound-based clinicoradiomics machine learning model.","authors":"Dong-Yue Wen, Jia-Min Chen, Zhi-Ping Tang, Jin-Shu Pang, Qiong Qin, Lu Zhang, Yun He, Hong Yang","doi":"10.1186/s12938-024-01259-3","DOIUrl":"10.1186/s12938-024-01259-3","url":null,"abstract":"<p><strong>Objectives: </strong>This study was designed to explore and validate the value of different machine learning models based on ultrasound image-omics features in the preoperative diagnosis of lymph node metastasis in pancreatic cancer (PC).</p><p><strong>Methods: </strong>This research involved 189 individuals diagnosed with PC confirmed by surgical pathology (training cohort: n = 151; test cohort: n = 38), including 50 cases of lymph node metastasis. Image-omics features were extracted from ultrasound images. After dimensionality reduction and screening, eight machine learning algorithms, including logistic regression (LR), support vector machine (SVM), K-nearest neighbors (KNN), random forest (RF), extra trees (ET), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and multilayer perceptron (MLP), were used to establish image-omics models to predict lymph node metastasis in PC. The best omics prediction model was selected through ROC curve analysis. Machine learning models were used to analyze clinical features and determine variables to establish a clinical model. A combined model was constructed by combining ultrasound image-omics and clinical features. Decision curve analysis (DCA) and a nomogram were used to evaluate the clinical application value of the model.</p><p><strong>Results: </strong>A total of 1561 image-omics features were extracted from ultrasound images. 15 valuable image-omics features were determined by regularization, dimension reduction, and algorithm selection. In the image-omics model, the LR model showed higher prediction efficiency and robustness, with an area under the ROC curve (AUC) of 0.773 in the training set and an AUC of 0.850 in the test set. The clinical model constructed by the boundary of lesions in ultrasound images and the clinical feature CA199 (AUC = 0.875). The combined model had the best prediction performance, with an AUC of 0.872 in the training set and 0.918 in the test set. The combined model showed better clinical benefit according to DCA, and the nomogram score provided clinical prediction solutions.</p><p><strong>Conclusion: </strong>The combined model established with clinical features has good diagnostic ability and can be used to predict lymph node metastasis in patients with PC. It is expected to provide an effective noninvasive method for clinical decision-making, thereby improving the diagnosis and treatment of PC.</p>","PeriodicalId":8927,"journal":{"name":"BioMedical Engineering OnLine","volume":"23 1","pages":"56"},"PeriodicalIF":2.9,"publicationDate":"2024-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11184715/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141417568","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}