Detection of left ventricular hypertrophy on the ECG through machine learning with a focus on obesity

Background

Cardiac remodeling, an important aspect of cardiovascular disease (CVD) progression, is emerging as a significant therapeutic target. The electrocardiogram (ECG) is of paramount importance in the initial evaluation of a patient. However, the ECG is not a sensitive method of detecting left ventricular hypertrophy (LVH), and as far as we know, it cannot detect changes in left ventricular geometry (LVG) at early stages, especially before LVH is present. Its sensitivity is particularly low for obese patients.

Purpose

To use a machine learning (ML) classifier to detect abnormal LVG from ECG parameters/markers, even before it becomes LVH, and to propose some indicative markers useful for practitioners. We also looked at the results of our model for obese patients to test the markers in this population.

Methods

We enrolled consecutive subjects, aged 30 years or older (mean age: 61.6±12 years old) with and without essential hypertension and no indications of CVD. All patients underwent a full echocardiographic evaluation and were classified into 2 groups; those with normal geometry (NG) vs. those with concentric remodeling (CR) or LVH defined as concentric hypertrophy (CH) and eccentric hypertrophy (EH). Abnormal LVG was identified as increased relative wall thickness (RWT) and/or left ventricular mass index (LVMi). We analyzed the EKG waveforms deduced to single beat averages for each lead using custom software and extracted 70 markers. We then trained a Random Forest machine learning model to classify subjects with abnormal LVG and calculated SHAP values to perform feature importance and interaction.

Results

After screening 1120 individuals, we enrolled 594 subjects, aged 30 years or older (mean age: 61.6±12 years old). The percentage of women was 56.5%, while 71.3% of all patients were hypertensive. Hypertension, age, body mass index divided by the Sokolow-Lyon voltage (BMI/S-L), QRS-T angle, and QTc duration were among the most important parameters (Figure, left panel) identified by the model as being predictive of abnormal LVG (AUC/ROC = 0.84, sensitivity = 0.94, specificity 0.61). Specifically for obese patients, whose prevalence in our population was 60.3%, our model performed well (sensitivity = 0.71, specificity = 0.92. When we tried our model without the the BMI/S-L parameter, the specificity dropped to 0.88. We also found that a cut-off point of 18 for the BMI/S-L marker predicted the patients who were more probable to have developed abnormal LVG.

Conclusions

This study is the first to demonstrate the promising potential of ML modeling for the efficient and cost-effective diagnostic screening of abnormal LVG and cardiac remodeling through ECG. We found specific clinical and ECG parameters that can predict early pathological changes of LVG in patients without established CVD and detect the population who will benefit from a detailed echocardiographic evaluation.

Funding Acknowledgement

Type of funding sources: None.

Artificial intelligence-based opportunistic screening for the detection of arterial hypertension through ECG signals

Background

Hypertension is a major risk factor for cardiovascular disease (CVD) which often escapes the diagnosis or should be confirmed by several office visits. The electrocardiogram (ECG) is one of the most widely used diagnostic tools and could be of paramount importance in patients' initial evaluation.

Purpose

To detect whether a person is hypertensive using features from the ECG, as well as basic anthropometric features such as age, sex, and body mass index (BMI).

Methods

We used machine learning (ML) techniques based features derived from the electrocardiogram for detecting hypertension in a population without CVD. We enrolled 1091 subjects who were classified into hypertensive and normotensive group. We trained 3 ML models, specifically logistic regression, k-nearest-neighbors, and random forest (RF), to predict the existence of hypertension in patients based only on a few basic clinical parameters and ECG-derived features. We also calculated Shapley additive explanations (SHAP), a sophisticated feature importance analysis, to interpret each feature's role in the random forest's predictions.

Results

Our RF model was able to distinguish hypertensive from normotensive patients with accuracy 84.2%, specificity 66.7%, sensitivity 91.4%, and area under the receiver-operating curve 0.86. Age, BMI, BMI-adjusted Cornell criteria (BMI multiplied by RaVL+SV3), R wave amplitude in aVL, and BMI-modified Sokolow-Lyon voltage (BMI divided by SV1+RV5), were the most important anthropometric and ECG-derived features in terms of the success of our model. Figure 1 shows the results in detecting hypertension by the Random Forest.

Conclusions

Our ML algorithm is effective in the detection of hypertension in patients using ECG-derived and basic anthropometric criteria. Our findings open new horizon in the detection of many undiagnosed hypertensive individuals who have an increased cardiovascular disease risk.

Funding Acknowledgement

Type of funding sources: None.

Prediction of abnormal left ventricular geometry on the ECG through machine learning with a focus on obesity

Background

Cardiac remodeling, an important aspect of cardiovascular disease (CVD) progression, is emerging as a significant therapeutic target. However, the ECG is not a sensitive method of detecting left ventricular hypertrophy (LVH), and as far as we know, it cannot detect changes in left ventricular geometry (LVG) at early stages, especially before LVH is present. Its sensitivity is particularly low for obese patients.

Purpose

To use a machine learning (ML) classifier to detect abnormal LVG from EKG parameters/markers, even before it becomes LVH, and to propose some indicative markers useful for practitioners. We also looked at the results of our model for obese patients to test the markers in this population.

Methods

We enrolled 594 consecutive subjects, aged 30 years or older (mean age: 61.6±12 years old) with and without essential hypertension and no indications of CVD. We tried to build a “clean” dataset through which we can target the clinical, anthropometric, and electrocardiogram measurements indicative of abnormal LVG. All patients underwent a full echocardiographic evaluation and were classified into 2 groups; those with normal geometry (NG) vs. those with concentric remodeling (CR) or LVH. Abnormal LVG was identified as increased relative wall thickness (RWT) and/or left ventricular mass index (LVMi). We analyzed the EKG waveforms deduced to single beat averages for each lead using custom software and extracted 70 markers. We then trained a Random Forest machine learning model to classify subjects with abnormal LVG and calculated SHAP values to perform feature importance and interaction.

Results

The percentage of women was 56.5%, while 71.3% of all patients were hypertensive. Hypertension, age, body mass index divided by the Sokolow-Lyon voltage (BMI/S-L), QRS-T angle, and QTc duration were among the most important parameters (Figure, left panel) identified by the model as being predictive of abnormal LVG (AUC/ROC = 0.84, sensitivity = 0.94, specificity 0.61). Specifically for obese patients, whose prevalence in our population was 60.3%, our model performed well (sensitivity = 0.71, specificity = 0.92. When we tried our model without the the BMI/S-L parameter, the specificity dropped to 0.88. We also found that a cut-off point of 18 for the BMI/S-L marker predicted the patients who were more probable to have developed abnormal LVG (Figure 1).

Conclusions

This study is the first to demonstrate the promising potential of ML modeling for the efficient and cost-effective diagnostic screening of abnormal LVG through ECG. We found specific clinical and ECG parameters that can predict early pathological changes of LVG in patients without established CVD and detect the population who will benefit from a detailed echocardiographic evaluation. Our model contributes to the development of human-centered and autonomous technologies and can optimize patient-management and treatment.

Funding Acknowledgement

Type of funding sources: None. Figure 1

Axitinib as a third or further line of treatment in renal cancer: a single institution experience

BACKGROUND

Kidney cancer is a lethal neoplasm that affects several thousands of people every year. Renal cell carcinoma (RCC) is the most common histologic type. Recent developments in the therapeutic approach include antiangiogenic targeted approaches and Immunotherapy. Thus, the therapeutic algorithm of RCC patients and the survival outcomes have changed dramatically.

METHODS

Herein we present a retrospective study of the patients treated in our Department with an antiangiogenic agent -Axitinib, a tyrosine kinase inhibitor- as a third or further line treatment. Statistical analysis was performed with SPSS, including the available clinicopathological data of the patients included.

RESULTS

Axitinib was found to be active in patients who received this treatment beyond second line. The toxicity profile of this regimen did not reveal any unknown adverse events.

CONCLUSIONS

Our real world data reflect that axitinib is a safe and effective option, even beyond the second line.

Patterns of practice and pharmacoeconomic analysis of the management of patients with metastatic renal cell carcinoma (mRCC) in Greece--the CRISIS study. A retrospective analysis by the Hellenic Genitourinary Cancer Group (HGUCG)

Background:Metastatic RCC (mRCC) treatment has been revolutionized with 11 approved targeted agents. We report patterns of practice, outcomes and pharmacoeconomic analyses after the introduction of targeted therapy. Patients and methods: CRISIS was a retrospective multicenter study of mRCCpatients who received targeted therapy . Results were related to the start of 1st-line therapy, with a cut off at 1 January 2011 in order to depict the impact of increased availability of effective options. Results: 164 patients, were included. 70.1% and 44.5% received 2nd and 3rd-line therapy, respectively. More patients were treated in 2nd-line after 1 January 2011. After a median follow-up of 55.1 months, median progression-free (PFS) and overall survival (OS) were 10.7 (95% confidence intervals [CI]: 8.3-13.7), 7.3 (95% CI: 5.1-8.6), 5.8 (95% CI: 3.8-7.8) and 34 (95% CI: 28.5-39.8), 22.4 (95% CI: 16-32.1), 18.3 (95% CI: 12.4-26.4) months for first, second and third line, respectively. Efficacy of sunitinib and pazopanib in 1st-line were similar. The mean total cost/patient was 35,012.2 Euros (standard deviation [SD]: 28,971.5). Conclusions: Our study confirms previous real-world data suggesting that continuing advances in the treatment of mRCC produce favorable outcomes in everyday practice. Pharmacoeconomic analyses are important for cost-effective utilization of emerging novel therapies.

[Overdentures (a case report)]

In this article, the advantages, disadvantages, indications, contraindications, and the details of overdenture treatment are outlined and a case is presented. We prepared an overdenture for a patient who came to the Clinic of Removable Prosthodontics of the University of Athens. The patient had a severe abrasion of his teeth as well as reduced vertical dimension (2.5 cm). The construction was made in such a way that we obtained a good esthetic results with no great expense, as the financial condition of the patient was the factor that influenced the plan of treatment basically.

[A clinical and epidemiological study of Tori mandibularis]

This is a study aiming at: a) reviewing the information found in the relevant literature as regards the etiology, incidence, distribution, implication and management of the tori mandibularis, b) evaluating the incidence, location and morphology of this bony mass in a sample of Greek population and c) comparing the findings with those of other investigators. The material consisted of 357 patients, from 20 years old and onwards, who had consecutively visited the Clinic of Removable Prosthodontics for some problem. After clinical examination and tabulation of the findings, the following conclusions were drawn: a) the etiology of appearance of tori mandibularis remains unknown; b) in our sample, 12.8% had this condition; c) the incidence was higher in men (60.4%) than in women (39.5%); d) in our sample, the higher percentage of individuals showing the condition originated from Thraci (Northern Greece) while the lower came from Hepiros; e) No indication of a heredity factor was found; f) this condition is more often bilateral than unilateral and g) the torus mandibularis was extending from the canine to the area of the first premolar in 54.4% of the cases.