Perfect prosthetic heart valve: generative design with machine learning, modeling, and optimization

Majority of modern techniques for creating and optimizing the geometry of medical devices are based on a combination of computer-aided designs and the utility of the finite element method This approach, however, is limited by the number of geometries that can be investigated and by the time required for design optimization. To address this issue, we propose a generative design approach that combines machine learning (ML) methods and optimization algorithms. We evaluate eight different machine learning methods, including decision tree-based and boosting algorithms, neural networks, and ensembles. For optimal design, we investigate six state-of-the-art optimization algorithms, including Random Search, Tree-structured Parzen Estimator, CMA-ES-based algorithm, Nondominated Sorting Genetic Algorithm, Multiobjective Tree-structured Parzen Estimator, and Quasi-Monte Carlo Algorithm. In our study, we apply the proposed approach to study the generative design of a prosthetic heart valve (PHV). The design constraints of the prosthetic heart valve, including spatial requirements, materials, and manufacturing methods, are used as inputs, and the proposed approach produces a final design and a corresponding score to determine if the design is effective. Extensive testing leads to the conclusion that utilizing a combination of ensemble methods in conjunction with a Tree-structured Parzen Estimator or a Nondominated Sorting Genetic Algorithm is the most effective method in generating new designs with a relatively low error rate. Specifically, the Mean Absolute Percentage Error was found to be 11.8% and 10.2% for lumen and peak stress prediction respectively. Furthermore, it was observed that both optimization techniques result in design scores of approximately 95%. From both a scientific and applied perspective, this approach aims to select the most efficient geometry with given input parameters, which can then be prototyped and used for subsequent in vitro experiments. By proposing this approach, we believe it will replace or complement CAD-FEM-based modeling, thereby accelerating the design process and finding better designs within given constraints. The repository, which contains the essential components of the study, including curated source code, dataset, and trained models, is publicly available at https://github.com/ViacheslavDanilov/generative_design.

Biomaterials Based on Carbon Nanotube Nanocomposites of Poly(styrene-b-isobutylene-b-styrene): The Effect of Nanotube Content on the Mechanical Properties, Biocompatibility and Hemocompatibility

Nanocomposites based on poly(styrene-block-isobutylene-block-styrene) (SIBS) and single-walled carbon nanotubes (CNTs) were prepared and characterized in terms of tensile strength as well as bio- and hemocompatibility. It was shown that modification of CNTs using dodecylamine (DDA), featured by a long non-polar alkane chain, provided much better dispersion of nanotubes in SIBS as compared to unmodified CNTs. As a result of such modification, the tensile strength of the nanocomposite based on SIBS with low molecular weight (M_n = 40,000 g mol^-1) containing 4% of functionalized CNTs was increased up to 5.51 ± 0.50 MPa in comparison with composites with unmodified CNTs (3.81 ± 0.11 MPa). However, the addition of CNTs had no significant effect on SIBS with high molecular weight (M_n~70,000 g mol^-1) with ultimate tensile stress of pure polymer of 11.62 MPa and 14.45 MPa in case of its modification with 1 wt% of CNT-DDA. Enhanced biocompatibility of nanocomposites as compared to neat SIBS has been demonstrated in experiment with EA.hy 926 cells. However, the platelet aggregation observed at high CNT concentrations can cause thrombosis. Therefore, SIBS with higher molecular weight (M_n~70,000 g mol^-1) reinforced by 1-2 wt% of CNTs is the most promising material for the development of cardiovascular implants such as heart valve prostheses.

Indirect supervision applied to COVID-19 and pneumonia classification

The novel coronavirus 19 (COVID-19) continues to have a devastating effect around the globe, leading many scientists and clinicians to actively seek to develop new techniques to assist with the tackling of this disease. Modern machine learning methods have shown promise in their adoption to assist the healthcare industry through their data and analytics-driven decision making, inspiring researchers to develop new angles to fight the virus. In this paper, we aim to develop a CNN-based method for the detection of COVID-19 by utilizing patients' chest X-ray images. Developing upon the inclusion of convolutional units, the proposed method makes use of indirect supervision based on Grad-CAM. This technique is used in the training process where Grad-CAM's attention heatmaps support the network's predictions. Despite recent progress, scarcity of data has thus far limited the development of a robust solution. We extend upon existing work by combining publicly available data across 5 different sources and carefully annotate the comprising images across three categories: normal, pneumonia, and COVID-19. To achieve a high classification accuracy, we propose a training pipeline based on indirect supervision of traditional classification networks, where the guidance is directed by an external algorithm. With this method, we observed that the widely used, standard networks can achieve an accuracy comparable to tailor-made models, specifically for COVID-19, with one network in particular, VGG-16, outperforming the best of the tailor-made models.

Aortography Keypoint Tracking for Transcatheter Aortic Valve Implantation Based on Multi-Task Learning

Currently, transcatheter aortic valve implantation (TAVI) represents the most efficient treatment option for patients with aortic stenosis, yet its clinical outcomes largely depend on the accuracy of valve positioning that is frequently complicated when routine imaging modalities are applied. Therefore, existing limitations of perioperative imaging underscore the need for the development of novel visual assistance systems enabling accurate procedures. In this paper, we propose an original multi-task learning-based algorithm for tracking the location of anatomical landmarks and labeling critical keypoints on both aortic valve and delivery system during TAVI. In order to optimize the speed and precision of labeling, we designed nine neural networks and then tested them to predict 11 keypoints of interest. These models were based on a variety of neural network architectures, namely MobileNet V2, ResNet V2, Inception V3, Inception ResNet V2 and EfficientNet B5. During training and validation, ResNet V2 and MobileNet V2 architectures showed the best prediction accuracy/time ratio, predicting keypoint labels and coordinates with 97/96% accuracy and 4.7/5.6% mean absolute error, respectively. Our study provides evidence that neural networks with these architectures are capable to perform real-time predictions of aortic valve and delivery system location, thereby contributing to the proper valve positioning during TAVI.

Two-color interferometer for study of dense low-ionized plasma on the target in high-power pulse linear accelerator

A two-color infrared interferometer has been developed for the investigation of high-density weakly ionized tantalum plasma in x-ray complexes based on linear induction accelerators (1.6 kA electron beam current, 4.6 MeV energy, and 100 ns pulse duration). Simultaneous probing at two different wavelengths makes it possible to independently measure the density of neutral and electron components. The interferometer uses wavelength values of 1.064 µm (Nd:YAG laser) and 10.6 µm (CO₂ laser). To attenuate the effect of sample beam refraction in inhomogeneous plasma, the interferometer used a refraction suppression scheme composed of spherical mirrors focusing the object beam into the region occupied by the plasma. In addition, the power of the sample beam transmitted through the plasma was controlled in order to analyze whether there was a distortion of the interference pattern because of strong sample beam refraction and absorption in the plasma cloud. To calibrate the initial phase shift of the probe radiation, a movable mirror mounted on a piezoelectric element and oscillating according to a harmonic law with amplitude greater than the laser wavelengths was used. In initial experiments, the parameters of target plasma registered by this interferometer are as follows: the linear density of neutrals reached 1.5 · 10¹⁷ cm^-2, and the degree of ionization was of the order of 10^-2. The target plasma expansion velocity is determined as ∼6 km/s.

Real-time coronary artery stenosis detection based on modern neural networks

Invasive coronary angiography remains the gold standard for diagnosing coronary artery disease, which may be complicated by both, patient-specific anatomy and image quality. Deep learning techniques aimed at detecting coronary artery stenoses may facilitate the diagnosis. However, previous studies have failed to achieve superior accuracy and performance for real-time labeling. Our study is aimed at confirming the feasibility of real-time coronary artery stenosis detection using deep learning methods. To reach this goal we trained and tested eight promising detectors based on different neural network architectures (MobileNet, ResNet-50, ResNet-101, Inception ResNet, NASNet) using clinical angiography data of 100 patients. Three neural networks have demonstrated superior results. The network based on Faster-RCNN Inception ResNet V2 is the most accurate and it achieved the mean Average Precision of 0.95, F1-score 0.96 and the slowest prediction rate of 3 fps on the validation subset. The relatively lightweight SSD MobileNet V2 network proved itself as the fastest one with a low mAP of 0.83, F1-score of 0.80 and a mean prediction rate of 38 fps. The model based on RFCN ResNet-101 V2 has demonstrated an optimal accuracy-to-speed ratio. Its mAP makes up 0.94, F1-score 0.96 while the prediction speed is 10 fps. The resultant performance-accuracy balance of the modern neural networks has confirmed the feasibility of real-time coronary artery stenosis detection supporting the decision-making process of the Heart Team interpreting coronary angiography findings.

[Minimally invasive thoracoscopic and robot-assisted lobectomy]

OBJECTIVE

To compare the early outcomes of video-assisted and robot-assisted lobectomy.

MATERIAL AND METHODS

There were 74 robot-assisted lung resections for the period 2017-2018. Sixty-six patients underwent lobectomy and 8 - segmentectomy. Patients were divided into 2 groups to compare thoracoscopic procedures. The main group consisted of 66 patients after robot-assisted lobectomy. The control group included 247 patients after video-assisted lobectomy. Mean age of patients was 60.5 years (range 35-70). There 72% men and 28% women.

CONCLUSION

Both thoracoscopic procedures are characterized by high efficiency and safety. However, certain disadvantages are peculiar to both methods.

Efficient workflow for automatic segmentation of the right heart based on 2D echocardiography

The present study aimed to present a workflow algorithm for automatic processing of 2D echocardiography images. The workflow was based on several sequential steps. For each step, we compared different approaches. Epicardial 2D echocardiography datasets were acquired during various open-chest beating-heart surgical procedures in three porcine hearts. We proposed a metric called the global index that is a weighted average of several accuracy coefficients, indices and the mean processing time. This metric allows the estimation of the speed and accuracy for processing each image. The global index ranges from 0 to 1, which facilitates comparison between different approaches. The second step involved comparison among filtering, sharpening and segmentation techniques. During the noise reduction step, we compared the median filter, total variation filter, bilateral filter, curvature flow filter, non-local means filter and mean shift filter. To clarify the endocardium borders of the right heart, we used the linear sharpen. Lastly, we applied watershed segmentation, clusterisation, region-growing, morphological segmentation, image foresting segmentation and isoline delineation. We assessed all the techniques and identified the most appropriate workflow for echocardiography image segmentation of the right heart. For successful processing and segmentation of echocardiography images with minimal error, we found that the workflow should include the total variation filter/bilateral filter, linear sharpen technique, isoline delineation/region-growing segmentation and morphological post-processing. We presented an efficient and accurate workflow for the precise diagnosis of cardiovascular diseases. We introduced the global index metric for image pre-processing and segmentation estimation.

[Video-assisted thoracoscopic thymectomy in myasthenia gravis. New approach]

AIM

To present possibility and technical aspects of anesthetic management during video-assisted thoracoscopic thymectomy in patients with myasthenia gravis.

MATERIAL AND METHODS

The article describes the proposed modified anesthetic method that is suitable for video-assisted thoracoscopic thymectomy in patients with myasthenia gravis. Nine patients with myasthenia underwent VATS-thymectomy. Anesthesia was performed with artificial airway and auxiliary ventilation without muscle relaxants administration.

[Anesthesia during video-assisted thoracoscopic lobectomy using laryngeal ductwork and auxiliary mechanical ventilation]

The article presents the experience of video-assisted thoracoscopic lobectomies in patients without endotracheal intubation and sparing of spontaneous breathing. The results of intraoperative monitoring and laboratory data of all perioperative period are described. In view of these data it was concluded that such approach can decline the severity of stress-response against surgical aggression in this group of patients.

[Anesthetic management of reconstructive surgery on the respiratory tract]

In the article provides an analysis of own of clinical experience peculiarities of carrying out of anesthesia, when performing 466 various tracheo-bronhoplastiс of operations for the period from 1979 to 2014 the year a one team of surgeons. Preoperative preparation, anesthesia and postoperative management of patients are key to improving treatment outcomes. Full arsenal of modern methods of ventilation lets you choose for each stage of surgery is most convenient for the surgeon and patient-safe way to maintain gas exchange. Presented in this paper our anestiology experience allows you to discuss and elect the proposed management of patients through the selection and optimization of the method that best meets modern requirements.

[Fluorescense enhancing of 5, 10, 15, 20-tetraphenylporphyrin in complex with human serum albumin and gold nanorods]

Using absorption and fluorescence spectroscopy methods we obtained the results demonstrating alterations in spectral characteristics in supramolecular system composed of gold nanorods (AuNR) (10 x 38 nm) and complexes of human serum albumin (HSA) and 5, 10, 15, 20-tetraphenylporphyrin (TPP). TPP fluorescence (max = 636 and 658 nm) was found to enhance. The dependence of fluorescence enhancing in time was of nonlinear nature. Maximum TPP fluorescence enhancing value was as high as 16% and it was achieved in 7 min after mixing the components. Simultaneously with TPP fluorescence enhancing we observed a decrease in HAS own fluorescence (max = 340 nm) and optical density reduction in maximum of longitudinal localized plasmon band of AuNR (max = 752 nm).

[Clinical and urodynamic evaluation of efficacy of conservative treatment of urination disorders in patients with benign prostate hyperplasia]

17 patients (mean age, 61 +/- 1.7 years) with an benign prostate hyperplasia (BPH) and urination disorders were under observation. After examination by ultrasound, urodynamic and laboratory studies, patients received conservative combined therapy, including alpha1-adrenoblocker silodosin, L-carnitine, B vitamins and picamilon. As a result of the 4,5-month treatment, most of the symptoms were eliminated as evidenced by the decrease in IPSS score from 17 to 3; urodynamic parameters have improved that was recorded by the home uroflowmetry. It was concluded that the use of proposed combined therapy in patients with BPH as the first stage of treatment, followed by the administration of 5alpha-reductase inhibitors or performing surgical correction in case of formation of bladder outlet obstruction is justified.

Electrically controlled polarized photoluminescence of CdSe/ZnS nanorods embedded in a liquid crystal template

A novel homogeneous composite material, consisting of luminescent CdSe/ZnS quantum nanorods, embedded in the nematic liquid crystal 5CB, has been prepared. Liquid crystal cells and free-standing stretched polymer films incorporating this composite material were characterized using polarized micro-photoluminescence and electro-optical measurements under an applied electric field. A liquid crystal induced, macroscopic orientation of the nanorods in a thin layer of the material has been demonstrated. A conventional liquid crystal cell, filled with this composite, exhibits 40% modulation of the nanorod's photoluminescence intensity when subjected to an external electric field. These results indicate that quantum nanorods may have practical applications in photonic devices.

[Pharmacourodynamic non-invasive studies in patients with prostatic adenoma]

After examination 47 patients with prostatic adenoma (mean age 63 years) received a course of alpha1-adenoblocker (omsulosin in a dose 0.4 mg/day). A three-day uroflowmetric monitoring was made at home before and after treatment. The findings were assessed with Liverpool nomo-gram by maximal flow. Omsulosin treatment was found to change voiding structure, to identify the obstructive component due to displacement of the flow characteristics to the low centiles of the nomogram. Thus, long-term treatment with alpha1-adrenoblockers can be used as a pharmacourodynamic test to detect infravesical obstruction in patients with prostatic adenoma.

[Sexual dysfunction in women with overactive bladder and their correction with m-cholinolytic spasmex]

A 4-month course of spasmex (15 mg three times a day) was conducted in 57 females (age 18-69 years, mean age 48 years) suffering from imperative urination manifesting as pollakiuria, imperative urges and urgent urinary incontinence in combination with sexual disorders. After the treatment the score of urination disorders went down from 21.4 to 12.7. Quality of life score decreased from 4.3 to 1.7. Sexual dysfunction diminished from 3.6 to 0.8 points. Clinical data showed that spasmex has a marked effect on m-cholinoreceptors, is well tolerated, has a moderate spasmolytic effect on the smooth muscles of the lower urinary tract. Moreover, spasmex can be used for correction of female sexual dysfunction caused by overactive bladder.

[Clinical and urodynamic rationale of the overactive bladder neurophysiological model]

The analysis of 58 cases of overactive bladder has shown that detrusor activity is not linked with clinical symptoms but is caused by supra segmentary lesion of the nervous system. The clinical picture of overactive bladder fits the proposed neurophysiological model where ischemic damage of the association neuron located in the lumbar spinal marrow triggers miction disorders leading to preganglionary detrusor denervation. Combined therapy with alpha1-adrenoblocker and M-cholinolytic is pathogenetically sound. Trospium chloride (spasmex) was used for long-term treatment of the patients as it produces minimal number of side effects and enables dose and schedule variation.

[Tadenan treatment of prostatic adenoma]

Tadenan, a drug of plant origin, was given to 27 patients with documented prostatic adenoma for 3-12 months. Home uroflowmonitoring for 1, 3, 6 and 12 months registered positive changes in urinary flow rate and urination volume. Side effects were absent. IPSS and QoL parameters were also assessed.

[24-h dynamics of urination in patients with overactive urinary bladder]

Home 3-day uroflowmetry was made in 41 female patients (mean age 46 years) before and after therapy for overactive bladder (OB). The course of therapy lasting from 6 weeks to 3 months included M-cholinolytics and vascular medication. Estimation of the bladder capacity in different time of 24-h period revealed circadian deviations of urinary flow and volumes. Administration of short-term M-cholinolytics enables chronotherapy providing maximal effects within 4-6 hours, possibility to vary doses. In view of this, optimal cholinergic therapy of OB is trospium chloride which has necessary properties, high safety and wide dose range.

[Changes in lower urinary tract function in children with hyperactive urinary bladder treated with combined therapy]

Low clinical efficacy of M-cholinolytics remains a serious problem in conservative therapy of hyperactive bladder (HB). To improve treatment outcomes, combined therapy (pikamilon plus driptan) was tried in 22 boys and 25 girls aged 7 to 14 years. The children were examined with uroflowmetry, cystometry. The response to the treatment was found to depend on mean effective capacity of the urinary bladder. Duration ofurodynamic correction has a negative colleration with this capacity. Urodynamic investigations performed in different periods of therapy show that improvement of lower urinary tract function is pathogenetically important for treatment of children with urologic pathology running in the presence of hyperactive urinary bladder.

[M-cholinolytic spasmex and its role in preoperative preparation of patients with urine incontinence]

Non-invasive uroflowmetric monitoring in females with combined incontinence has demonstrated that hyperactive urinary bladder and stress urine incontinence in females are similar in relation to functional disorders of the low urinary tract both in complete and partial syndrome of imperative urination. In the latter case, when imperative urine incontinence is absent, voiding structure is changed in the same manner, and the patients' preparation for surgical intervention demands long-term treatment with M-cholinolytics. We propose to use spasmex (trospium chloride) which has minimal side effects in long-term therapy. Conservative therapy of such patients with spasmex radically changes background for planning surgical intervention.

[Substantiation of the duration of m-cholinolytics administration in conservative treatment of hyperactive urinary bladder]

The analysis of treatment outcomes in 74 patients with hyperactive urinary bladder (HUB) has shown that duration of the treatment depends on the initial function of the lower urinary tracts. There was a correlation between mean effective therapy and treatment duration. Estimation of this correlation allows prognosis of treatment efficacy and selection of optimal anticholinergic drug for correction of miction disorders. Spasmex proved to be the most adequate drug for long-term treatment due to minimal number of side effects and feasibility to vary dose and regimen of intake.

[The role of alpha1-adrenoblocker alfuzosine in diagnostic-therapeutic algorithm of hyperactive urinary bladder management in women and selection of patients for treatment with M-cholinergic antagonists]

In management of patients with hyperactive urinary bladder (HAUB) it is necessary to take into consideration the fact that long-term intake of M-cholinergic antagonists may result in obstructive miction because of low contractile activity of the urinary bladder. Use of M-cholinergic antagonists also problematic in detruzor dysfunction leading to slowing down of the urinary flow. In view of these possible complications, an algorithm of urodynamic diagnosis is proposed based on non-invasive registration of miction disorders using home uroflowmetry. Correction of detruzor contractility with dalfas combines well with cholinoblockers. In some cases administration of nontitrated alphablockers has no alternative especially in long-term treatment with mediators suppressing nonadapted detruzor contractions as it happens in HAUB. The above diagnosticotherapeutic algorithm of female miction disorders management which regards urine flow rates and comprises intake of alfusozine if necessary allows treatment of females with small urinary bladder when cholinergic antagonists are low-effective.

[Combined treatment of voiding disorders in women with m-cholinolytic and alpha 1-adrenoblocker]

The drug spasmex (trospium chloride) having an m-cholinolytic and myotropic spasmolytic action has been used for treatment of enuresis in women with imperative voiding syndrome. Spasmex was given in combination with alpha 1-adrenoblockers to 20 women. The clinical picture was assessed with the original table. Urodynamic alterations were registered by means of home uroflowmeter. Clinical parameters improved 2 times after 4 weeks of the treatment. Mean effective capacity of the urinary bladder increased by 25%. Spasmex in combination with terasozine was well tolerated and had no side effects.

[Role of urodynamics correction in combined therapy of children with urinary infection]

The authors propose differentiated complexes of therapy of children with urinary infection. These complexes include medicines (M-cholinolytics, antihypoxic drugs and drugs improving microcirculation) and low-invasive therapy. Clinical experience is reported for 4 boys and 40 girls with microbial-inflammatory urinary diseases. Standard urodynamic tests were made in all the patients before and after the course of therapy. Three variants of driptan effects on the urinary bladder were identified: M-cholinolytic, spasmolytic and mixed (larger volume and lower intravesical hypertension). Some of the patients experienced standard endoscopic reflux correction with the use of synthetic hydrophilic gel. Control urodynamic examinations 3 months after the treatment showed that functional disorders were reduced. Treatment efficacy was assessed with home flowmetry. Thus, combined therapy of microbial-inflammatory diseases with parallel correction of lower urinary tracts urodynamics by M-cholinolytics improves treatment results. Courses of antibacterial and uroceptic therapy should be followed by non-invasive monitoring to reveal urodynamic disorders, define dysfunction type and perform an additional treatment with mediators in combination with physiotherapeutic procedures.

[Prospects for differentiated conservative treatment of patients with benign prostatic hyperplasia]

The results of conservative therapy of 30 patients aged from 50 to 79 years with benign hyperplasia of prostate according to the IPSS scale from 7 to 30 points were analyzed. Under the influence of treatment in all the patients the clinical symptoms severity, according to the monitoring data using the IPSS tables, had reduced by 72% at average and quality of life improve no less than by 50%. Average efficacy of the volume and volumetric of urination had enhanced according to uroflowmetry data.

Effects of resonances on halo formation in high-intensity storage rings

Numerical calculations for the Spallation Neutron Source accumulator ring indicate that lattice resonances excited by the space-charge potential can increase a mismatch significantly by deforming the beam distribution in phase space. Hence increased mismatch leads to enhanced envelope oscillations that are driving the 2:1 parametric resonance leading to halo formation, even for initially matched beams. We have observed this behavior for the 2 nu(x) - 2 nu(y) = 0 resonance and for the 4 nu(y) = 23 resonance. This mechanism for halo formation peculiar to rings through resonance driven mismatch is very sensitive to the tunes, which emphasizes the importance of a careful choice of operating point in tune space.