A New Multi-Agent Feature Wrapper Machine Learning Approach for Heart Disease Diagnosis

Heart disease(HD)is a serious widespread life-threatening disease.The heart of patients with HD fails to pump sufcient amounts of blood to the entire body.Diagnosing the occurrence of HD early and efciently may prevent the manifestation of the debilitating effects of this disease and aid in its effective treatment.Classical methods for diagnosing HD are sometimes unreliable and insufcient in analyzing the related symptoms.As an alternative,noninvasive medical procedures based on machine learning(ML)methods provide reliable HD diagnosis and efcient prediction of HD conditions.However,the existing models of automated ML-based HD diagnostic methods cannot satisfy clinical evaluation criteria because of their inability to recognize anomalies in extracted symptoms represented as classication features from patients with HD.In this study,we propose an automated heart disease diagnosis(AHDD)system that integrates a binary convolutional neural network(CNN)with a new multi-agent feature wrapper(MAFW)model.The MAFW model consists of four software agents that operate a genetic algorithm(GA),a support vector machine(SVM),and Naïve Bayes(NB).The agents instruct the GA to perform a global search on HD features and adjust the weights of SVM and BN during initial classication.A nal tuning to CNN is then performed to ensure that the best set of features are included in HD identication.The CNN consists of ve layers that categorize patients as healthy or with HD according to the analysis of optimized HD features.We evaluate the classication performance of the proposed AHDD system via 12 common ML techniques and conventional CNN models by using across-validation technique and by assessing six evaluation criteria.The AHDD system achieves the highest accuracy of 90.1%,whereas the other ML and conventional CNN models attain only 72.3%–83.8%accuracy on average.Therefore,the AHDD system proposed herein has the highest capability to identify patients with HD.This system can be used by medical practitioners to diagnose HD efciently。

Light-Activated Virtual Sensor Array with Machine Learning for Non-Invasive Diagnosis of Coronary Heart Disease

Early non-invasive diagnosis of coronary heart disease(CHD)is critical.However,it is challenging to achieve accurate CHD diagnosis via detecting breath.In this work,heterostructured complexes of black phosphorus(BP)and two-dimensional carbide and nitride(MXene)with high gas sensitivity and photo responsiveness were formulated using a self-assembly strategy.A light-activated virtual sensor array(LAVSA)based on BP/Ti_(3)C_(2)Tx was prepared under photomodulation and further assembled into an instant gas sensing platform(IGSP).In addition,a machine learning(ML)algorithm was introduced to help the IGSP detect and recognize the signals of breath samples to diagnose CHD.Due to the synergistic effect of BP and Ti_(3)C_(2)Tx as well as photo excitation,the synthesized heterostructured complexes exhibited higher performance than pristine Ti_(3)C_(2)Tx,with a response value 26%higher than that of pristine Ti_(3)C_(2)Tx.In addition,with the help of a pattern recognition algorithm,LAVSA successfully detected and identified 15 odor molecules affiliated with alcohols,ketones,aldehydes,esters,and acids.Meanwhile,with the assistance of ML,the IGSP achieved 69.2%accuracy in detecting the breath odor of 45 volunteers from healthy people and CHD patients.In conclusion,an immediate,low-cost,and accurate prototype was designed and fabricated for the noninvasive diagnosis of CHD,which provided a generalized solution for diagnosing other diseases and other more complex application scenarios.

Utilising ductus venosus Doppler waveform and four-chamber view to screen for foetal cardiac malformation in early second trimester of pregnancy

Background Foetal echocardiography has become a diagnostic method to detect foetal congenital heart disease with high probability. However, it is not only time consuming and but also difficult to visualize outflow tract of foetus early in the second trimester of pregnancy, even for an experienced obstetric uhrasonographer. Recently, many methods for screening foetal cardiac anomalies were explored, but much more work is needed to develop an effective and suitable screening method. The aim of this study was to investigate the clinical significance of utilising the ductus venosus （DV） Doppler examination and the four-chamber view of heart to screen for foetal cardiac malformation in early second trimester of pregnancy. Methods Heart and DV of 401 consecutive foetuses in early second trimester （12^＋1- 17^ ＋6 weeks） in high risk pregnancies were examined with Acuson 128 xp/10 or Sequoia 512 ultrasound diagnostic systems. Absent or reversed flow during atrial contraction （A-wave） in the DV was defined as sufficiently abnormal to screen for foetal cardiac malformations. The foetal echocardiographic diagnosis was confirmed by postnatal echocardiography （or postmortem）. The sensitivities of screening tests were compared among the three methods： DV Doppler examination, four-chamber view alone, and the combination of both techniques.Results Satisfactory examinations were obtained in 383/401 foetuses （95%）. Thirty foetuses with cardiac abnormalities were confirmed by neonatal echocardiography （ or postmortem ）. The sensitivity of DV Doppler examination or four-chamber view alone is 63 % （19/30） and 60 % （ 18/30）, respectively. The sensitivity of combining information, DV Doppler flow waveform and four-chamber view, to screen for foetal cardiac malformation is 83% （25/30） and significantly better than that of either DV Doppler flow waveform or four chamber view alone （ P 〈 0. 05 ）. Conclusion Doppler flow waveform of DV can be used to screen for foetal cardiac malformation early in the second trimester. Combining information from Doppler flow waveform of DV and four-chamber view will improve the overall sensitivity of the screening.