Doctor's Decision Research Articles

Do patients prefer a human doctor, artificial intelligence, or a blend, and is this preference dependent on medical discipline? Empirical evidence and implications for medical practice.

Today the doctor-patient relationship typically takes place in a face-to-face setting. However, with the advent of artificial intelligence (AI) systems, two further interaction scenarios are possible: an AI system supports the doctor's decision regarding diagnosis and/or treatment while interacting with the patient, or an AI system could even substitute the doctor and hence a patient interacts with a chatbot (i.e., a machine) alone. Against this background, we report on an online experiment in which we analyzed data from N = 1,183 people. The data was collected in German-speaking countries (Germany, Austria, Switzerland). The participants were asked to imagine they had been suffering from medical conditions of unknown origin for some time and that they were therefore visiting a health center to seek advice from a doctor. We developed descriptions of patient-doctor interactions (referred to as vignettes), thereby manipulating the patient's interaction partner: (i) human doctor, (ii) human doctor with an AI system, and (iii) an AI system only (i.e., chatbot). Furthermore, we manipulated medical discipline: (i) cardiology, (ii) orthopedics, (iii) dermatology, and (iv) psychiatry. Based on this 3 × 4 experimental within-subjects design, our results indicate that people prefer a human doctor, followed by a human doctor with an AI system, and an AI system alone came in last place. Specifically, based on these 12 hypothetical interaction situations, we found a significant main effect of a patient's interaction partner on trust, distrust, perceived privacy invasion, information disclosure, treatment adherence, and satisfaction. Moreover, perceptions of trust, distrust, and privacy invasion predicted information disclosure, treatment adherence, and satisfaction as a function of interaction partner and medical discipline. We found that the situation in psychiatry is different from the other three disciplines. Specifically, the six outcome variables differed strongly between psychiatry and the three other disciplines in the "human doctor with an AI system" condition, while this effect was not that strong in the other conditions (human doctor, chatbot). These findings have important implications for the use of AI in medical care and in the interaction between patients and their doctors.

A system for assessing the effect of stress on the physiological state of a patient with sleep apnea

Problem statement. In modern society, the problems associated with sleep apnea are manifested against the background of public health. One of the reasons for this is social stress. Prolonged social stress forms stable transcription profiles when sleep apnea occurs, which has consequences for the patient in the form of various diseases. At the present stage, science is trying to understand which diagnostically significant indicators are modeled depending on the type of stress in the diagnosis of sleep apnea. Goal. To develop a system for assessing the physiological state of a patient with AS, depending on the type of stress, to support a doctor's decision on the diagnosis of the disease. Results. The principles of synthesis of the stress type assessment system in the diagnosis of AS are considered. Based on them, a model of a random process for assessing the effect of stress on the physiological state of a patient with sleep apnea, a method for determining the type of stress in sleep apnea, the combination of which can be defined as a solution to the scientific problem of improving the quality of life of patients with sleep apnea, is proposed. Practical significance. A decision support system for a doctor in assessing the physiological condition of a patient with sleep apnea.

The Optimal Model for Copy-Move Forgery Detection in Medical Images.

Digital devices can easily forge medical images. Copy-move forgery detection (CMFD) in medical image has led to abuses in areas where access to advanced medical devices is unavailable. Forgery of the copy-move image directly affects the doctor's decision. The method discussed here is an optimal method for detecting medical image forgery. The proposed method is based on an evolutionary algorithm that can detect fake blocks well. In the first stage, the image is taken to the signal level with the help of a discrete cosine transform (DCT). It is then ready for segmentation by applying discrete wavelet transform (DWT). The low-low band of DWT, which has the most image properties, is divided into blocks. Each block is searched using the equilibrium optimization algorithm. The blocks are most likely to be selected, and the final image is generated. The proposed method was evaluated based on three criteria of precision, recall, and F1 and obtained 90.07%, 92.34%, and 91.56%, respectively. It is superior to the methods studied on medical images. It concluded that our method for CMFD in the medical images was more accurate.

A qualitative study on patients' selection in the scarcity of resources in the COVID‐19 pandemic in a communal culture

AbstractThe scarcity of resources during the COVID‐19 pandemic caused ethical dilemmas in prioritizing patients for treatment. Medical and ethical guidance only emphasizes clinical procedures but does not consider the sociocultural aspect. This study explored the perception of former COVID‐19 patients and their families on the decision‐making process of the patient's selection at a time of scarcity of resources. The result will inform the development of an ethical guide for allocating scarce resources that aligns with Indonesian culture. We conducted qualitative research with in‐depth interviews between May ‐ December 2022 involving sixteen participants from various cities in Indonesia. We transcribed the interviews and analyzed the results using thematic analysis. This study found that doctor's decisions often differed from patient's expectations in allocating scarce resources, and therefore, it should be communicated appropriately. Medical decisions were not sufficiently made ethically, but must also be made communicatively. In Indonesia's strong communal culture, community involvement was essential to distributing limited resources. A better approach to ethical education, including adequate communication skills, is necessary to prepare health professionals for facing unpredictable future pandemics.

Typical and atypical chest pain: an equivocal and dangerous definition

Abstract Background The traditional classification of stable chest pain into typical, atypical and non anginal1 is equivocal and can lead to doctor's decisions and patient's perception that do not correspond to the real importance of the clinical picture. Methods In the last 15 years, 3,588 outpatients, aged >35 years (mean 64±9, female 55%) were evaluated for stable chest pain reported for more than 48 hours. We excluded 290 patients with a history of myocardial infarction or angina pectoris, heart failure, valve diseases or any other major cardiovascular event, or ECG signs of recent ischemia. In the other 3,298 cases we applied the traditional classification of chest pain. All patients were evaluated for the presence of HEART score risk factors² (smoking, obesity, family history of cardiovascular events, hypertension, diabetes, hypercholesterolemia) as well as age. All patients with typical or atypical angina and those with non-anginal pain according to the physician's judgment were evaluated with one or more functional cardiovascular tests (exertion test, nuclear stress tests, stress echo).A follow-up was performed at 6 months and 1 year, for the detection of major cardiovascular events (cardiovascular death, non-fatal heart attack, revascularization, stroke) and non-cardiovascular mortality. Results Table 1 shows the results of chest pain classification and positive functional test. Table 2 shows major events at one year of follow-up stratified according to the initial assessment of chest pain. Table 3 stratifies the events according to the extent of the risk factors and age. It is evident that a positive functional test is also present in 10% of patients with atypical pain and in 4% of those with non-anginal pain. Patients with typical pain undergo revascularization more frequently (39%). The peculiarity of our study is that the prognosis and rate of events do not differ in terms of clinical and ischemic burden among the 3 groups at 1 year of follow-up, except when patients are stratified by risk burden and for age (Table 4). About 1% of patients had a clinical event regardless of the type of pain and the presence of ischemia and 87% of the events involved patients with ≥ 3 risk factors or heavy smokers. Nine percent of all heavy smokers aged <65 years had a major event and are the highest risk group with a peak of 12.5% in those with atypical pain. Conclusions The traditional classification of stable chest pain should be abandoned, as it does not correlate with actual disease burden and prognosis, that appears to be linked to the burden of cardiovascular risk and above all to smoking in patients < 65 years of age, rather than to the clinical and ischemic burden. It will be necessary to evaluate whether the anatomical burden for the search of high-risk coronary lesions improves the prognosis with added value with respect to the treatment of risk factors³.Picture 1Picture 2

An Automatized Deep Segmentation and Classification Model for Lumbar Disk Degeneration and Clarification of Its Impact on Clinical Decisions.

Cross-sectional database study. The purpose of this study was to develop a successful, reproducible, and reliable convolutional neural network (CNN) model capable of segmentation and classification for grading intervertebral disc degeneration (IVDD), as well as quantify the network's impact on doctors' clinical decision-making. 5685 discs from 1137 patients were graded separately by four experienced doctors according to the Pfirrmann classification. A ground truth (GT) was established for each disc in accordance with the decision of the majority of doctors. The U-net model is used for segmentation. 1815 discs from 363 patients were used to train and test the U-net. The Inception V3 model is employed for classification. All discs were separated into two distinct sets: 90% in a training set and 10% in a test set. The performance metrics of these models were measured. Reliability tests were performed. The impact of CNN assistance on doctors was assessed. Segmentation accuracy was .9597 with a .8717 Jaccard Index and a .9314 Sorensen Dice coefficient. Classification accuracy is .9346, and the F1 score is .9355. The intraclass correlation coefficient (ICC) and kappa values between CNN and GT were .95-.97. With CNN's assistance, the success rates of doctors increased by 7.9% to 22%. The fully automated network outperformed doctors markedly in terms of accuracy and reliability. The results of CNN were comparable to those of other recent studies in the literature. It was determined that CNN's assistance had a substantial positive effect on the doctor's decision.

ChatGPT-3.5 and ChatGPT-4 dermatological knowledge level based on the Specialty Certificate Examination in Dermatology.

The global use of artificial intelligence (AI) has the potential to revolutionize the healthcare industry. Despite the fact that AI is becoming more popular, there is still a lack of evidence on its use in dermatology. To determine the capacity of ChatGPT-3.5 and ChatGPT-4 to support dermatology knowledge and clinical decision-making in medical practice. Three Specialty Certificate Examination in Dermatology tests, in English and Polish, consisting of 120 single-best-answer, multiple-choice questions each, were used to assess the performance of ChatGPT-3.5 and ChatGPT-4. ChatGPT-4 exceeded the 60% pass rate in every performed test, with a minimum of 80% and 70% correct answers for the English and Polish versions, respectively. ChatGPT-4 performed significantly better on each exam (P < 0.01), regardless of language, compared with ChatGPT-3.5. Furthermore, ChatGPT-4 answered clinical picture-type questions with an average accuracy of 93.0% and 84.2% for questions in English and Polish, respectively. The difference between the tests in Polish and English were not significant; however, ChatGPT-3.5 and ChatGPT-4 performed better overall in English than in Polish by an average of 8 percentage points for each test. Incorrect ChatGPT answers were highly correlated with a lower difficulty index, denoting questions of higher difficulty in most of the tests (P < 0.05). The dermatology knowledge level of ChatGPT was high, and ChatGPT-4 performed significantly better than ChatGPT-3.5. Although the use of ChatGPT will not replace a doctor's final decision, physicians should support the development of AI in dermatology to raise the standards of medical care.

The use of antibacterial drugs in the treatment of coronavirus infection in children in real outpatient clinic practice

Introduction. In light of the available data on the unjustified use of antibiotics in children with acute respiratory viral infections, the frequency of prescribing antibacterial drugs to children with a new coronavirus infection in real clinical practice is of great practical and scientific interest.The aim of the study was to study the frequency of antibiotic therapy and identify factors that increase the risk of needing antibiotics in children with coronavirus infection at the outpatient stage.Materials and methods. A retrospective cross-sectional study was conducted on the basis of children's polyclinics in Samara from November 2021 to July 2022, which included a survey of parents and an analysis of outpatient records of 100 children who were under outpatient observation and treatment for COVID-19. An analysis was made of the frequency of prescribing antibiotics to them and potential factors predisposing to their prescribing to children with COVID-19. The obtained data were subjected to statistical processing using the IBM SPSS Statistics 25 program and a comparative analysis.Results. The frequency of use of antibiotic therapy according to the results of the study was 11 % and did not exceed the Russian average data (14.3 %). When analyzing the factors influencing the doctor's decision to prescribe an antibiotic to a child with coronavirus infection, the most important were a history of recurrent respiratory infections (children from the FIC group), as well as the presence of concomitant diseases and background conditions in children, primarily chronic bronchopulmonary diseases. and nervous system.Discussion. The prescription of antibiotics for children with a new coronavirus infection at the outpatient stage remains empirical due to the limitations of the use of laboratory and instrumental methods for additional examination of the child, which objectifies the likelihood of a secondary bacterial infection. Despite this, the frequency of antibiotic use according to the results of the study was 11 % and did not exceed the Russian average data (14.3 %), as well as the average frequency of antibiotic use for acute respiratory viral infections in children in the Russian Federation (21.5-23.3 %).

Behavioral Components and Context of Antimicrobial Prescription in a Tertiary Hospital in Portugal.

Consumption of antimicrobials is an important driver of antimicrobial resistance. There is limited knowledge of the key determinants of antimicrobial prescribing behavior in hospitals. An understanding of these determinants is required for the successful design, adoption, and implementation of quality improvement interventions in Antimicrobial Stewardship Programs (ASP). This study aimed to describe the main factors that influence the doctor's decision on antimicrobials prescribing and to identify the behaviors that drive physicians' decision making. A structured web-based questionnaire focused on behavioral components of antimicrobial prescription was applied to the medical staff of three different departments-Internal Medicine, General Surgery, and Intensive Care Medicine-of a university hospital. All doctors agreed that inadequate use of antimicrobials increases AMR. A total of 77% of the surgeons and 100% of the internists and intensivists perceived antimicrobial prescription as a priority in the department. Full autonomy in antimicrobial prescription was preferred by internists (64%) but not by surgeons (18%) and intensivists (24%). Most physicians were keen to have ASP advice, but most did not want advice from colleagues of the same service. Almost all surgeons ask for advice when prescribing, but only 68% of the internists do it. Less than half of all physicians and only 25% of the surgeons felt free to prescribe contrary to guidelines. Most physicians, particularly in Intensive Care Medicine (94%), adopt the "wait and see" strategy when no microbiologic confirmation is available, but 27% of the surgeons start empirical therapy. In conclusion, the context of antimicrobial prescription, autonomy, and confidence in antimicrobial prescription demonstrated heterogeneity between the three departments and this should be considered when planning ASP.

Real-world experience of rituximab biosimilar GP2013 in rheumatoid arthritis patients naïve to or switched from reference rituximab.

Randomized controlled trials showed high comparability of biosimilar rituximab (bs-RTX) GP2013 to biologic originator RTX (bo-RTX). Data on effectiveness of switching from bo-RTX to bs-RTX, starting therapy with bs-RTX, and bs-RTX drug survival in real-world setting are sparse. To explore long-term drug effectiveness and survival of bs-RTX GP2013 in rheumatoid arthritis (RA) patients both naïve to and mandatory switched from bo-RTX, and to clarify reasons for treatment cessation. Retrospective observational cohort study including RA outpatient clinic patients treated with bs-RTX between 2018 and 2021 in Norway. Patients were examined and monitored using recommended measures for disease activity and patient-reported outcomes (PROs). For description of population medians and interquartile range were used. Difference between observation times was assessed with Signed-Rank test, drug survival with Kaplan-Meier survival analysis. Reasons for discontinuation were ascertained. Among 110 patients, at baseline, 88 were mandatory switched from bo-RTX and 22 were RTX-naïve. During 2-year follow-up, disease activity and PROs measures remained stable in switchers subgroup and improved in subgroup starting bs-RTX for the first time. Overall drug survival was 80.0% after 1year and 57.7% after 2years and was significantly higher in bs-RTX-switched than in bs-RTX-naïve patients (p = 0.036). Two most frequently reported reasons for drug discontinuation were remission (38.6%) and doctor's decision (27.1%). RA patients treated with bs-RTX had satisfactory treatment response and drug retention rates which supports equivalence of bs-RTX GP2013 to bo-RTX, both in patients naïve to and mandatory switched from bo-RTX.

Применение сверточных нейронных сетей для определения эндотелиальной дистрофии Фукса

Purpose. To evaluate the application of convolutional neural networks for the automatic detection of Fuchs' dystrophy. Material and methods. The study included 700 biomicroscopic images of the corneal endothelium (Tomey EM-3000) randomly selected from the database of the Saint-Petersburg brunch of the S. Fyodorov Eye Microsurgery Federal State Institution. At the first stage, the images were divided into 2 groups. The first group included images with the presence of Fuchs' dystrophy, the second – another pathology or a healthy cornea. The corneal endothelial cell density images were divided into three categories: training, validation, and test datasets. In our study we tested various architectures of convolutional neural networks: ResNet18, ResNet50, VGG16, VGG19 and GoogleNet. Results. The approbation of the neural network on the test sample has given the following values of the F-measure: ResNet18: 0.985; ResNet50: 1,000; VGG16: 0.940; VGG19: 0.990; GoogleNet: 0.987. Pre-trained network ResNet50 performed best with frozen layers, Adam optimizer, cross-entropy as a loss function, and a training step of 0.000005. Conclusion. The use of convolutional neural networks for the automatic detection of Fuchs' dystrophy can be successfully implemented as part of a doctor's decision support system. ResNet50 showed the best results among all types of models and did not give a single error on the test sample, which indicates the high efficiency of using this network in the classification algorithm for corneal endothelial images. Keywords: artificial intelligence, Fuchs corneal dystrophy, convolutional neural networks

Efficacy and Safety of Standardized Protocol of CPAP Therapy for Full-Term Newborns in Delivery Room at Transient Tachypnea: Clinical Trial with Historical Control

Background. Nowadays there is no consensus on the tactics and optimal protocol for Continuous Positive Airway Pressure (СРАР) at transient tachypnea of newborns (TTN) in delivery room.Objective. The aim of the study is to examine the efficacy and safety of standardized protocol of CPAP therapy for newborns with TTN in delivery room.Methods. The clinical study (implementation of standardized CPAP protocol) included full-term infants (gestational age — 37–41 weeks) with diagnosed TTN and CPAP therapy during first 60 minutes of life. Similar inclusion criteria were applied for the historical control group (born within previous year for whom CPAP was implemented according to “usual” protocol). Initiation of mononasal CPAP in main group was carried out when respiratory disorders were assessed according to Downes scale ≥ 3 points with control points at 20–40–60 minutes via the same scale, in control group at ≥ 4 points — for all cases, and according to the doctor's decision at 1–3 points. Mean airway pressure was maintained at 8 and 5–10 cm H2O, CPAP duration was 20–60 and 5–30 min, respectively. The major endpoints: the frequency of patient transfer from delivery room to intensive care unit or hospitalization to the neonatal pathology department, as well as total period of hospitalization. Moreover, frequency of invasive manipulations (intravenous catheterization, parenteral feeding), antibacterial therapy, cerebral injuries (cerebral ischemia, intracerebral hemorrhage), nasal passages injuries, pneumothorax (in the first 24 hours of life) were recorded during the entire hospitalization period.Results. 140 newborns with TTN were included in the clinical study, 30 were excluded from the study, specifically 13 due to violation of the CPAP protocol. The historical control group included 165 newborns. This groups were comparable for most baseline (before the start of CPAP) indicators except for maternal COVID-19 frequency during pregnancy and twin newborns frequency. This groups were comparable for most baseline (before the start of CPAP) indicators except for the frequency of maternal COVID-19 cases during pregnancy and the frequency of twin newborns. Hospitalization rate in intensive care units (18.2 versus 70.3%; p &lt; 0.001) and neonatal pathology departments (31.8 versus 80.0%; p &lt; 0.001), as well as total period of hospitalization (3 versus 10 days; p &lt; 0.001) were lower in the standardized CPAP therapy group. Lower frequency of invasive manipulations, antibacterial therapy, and cerebral ischemia was recorded in this group. The safety of СРАР administration in delivery room was confirmed by the absence of nasal passages injuries in both groups, as well as comparable frequency of pneumothorax.Conclusion. The use of standardized CPAP protocol in delivery room for full-term newborns with TTN had higher rate of favorable hospitalization outcomes. Study limitations require validation of all the findings in independent studies.

Shared decision making for patients with COVID-19 in a public training hospital in Mashhad, Iran

ABSTRACT Objectives This cross-sectional study was conducted on 359 COVID-19 patients in a public training hospital in Mashhad to assess the level of shared decision-making among the patients with COVID-19 and determine the affecting factors. Shared decision making was evaluated with a questionnaire based on Iranian accreditation standards. Data were analyzed by employing Chi-square test and stepwise logistic. The models were calibrated by the Hosmer–Lemeshow test. Results The results show that maximum shared decision making does not occur among these patients. 74% of patients stated that they have participated in the doctor's decision for the treatment plan. The results showed that gender, educational status and employment status (had significant relationships with patient participation in physician decisions).

Exit, voice or neglect: Understanding the choices faced by doctors experiencing barriers to leading health system change through the case of Sierra Leone

This paper presents a study from Sierra Leone that explored the experiences of doctors as they endeavored to improve the health care systems in which they worked. Twenty-eight interviews were conducted with doctors in Sierra Leone, complemented by long-standing experience of national health provision and research by the authors. Drawing on Hirschman's theory of ‘exit, voice and loyalty’, the paper's framework analysis elaborates the doctor's career decisions and choices under systematic political and economic constraints, and in particular, the specter of retribution, including posting to undesirable jobs and withholding of salaries. This retribution was considered a driver of exit by doctors from the system, and few examples were given of doctors successfully advocating for change through advocacy (‘voice’). We suggest that the relevance of Hirschman's theory to this setting is in drawing attention to the critical themes of retribution, opportunity, loyalties, and partial exits, ones often neglected in efforts to reduce emigration of doctors and strengthen their leadership. Ultimately, this paper critiques the overemphasis of mechanistic ‘capacity building’ in global health and recommends that health system strengthening must be viewed as a jointly political as well as technical exercise.

Preferred Role in Health Care Decision Making Over Time in Patients With Heart Failure: My Decision or My Doctor's Decision?

Understanding patients' preferred role in decision making can improve patient-centered care. This study aimed to determine change and the predictors of change in preferred decision-making roles over time in patients with heart failure. During the CASA (Collaborative Care to Alleviate Symptoms and Adjust to Illness) trial, patients' preferred roles in decision making were measured using the Control Preferences Scale (range 1-5; higher = less active; n = 312) at 4 timepoints over 1 year. The effect of the CASA intervention on preferred decision-making roles was tested using generalized linear mixed models. Whether preferences changed over time in the whole population was determined using linear regression. Demographic and health-related factors were examined as predictors of change using multiple linear regression. At baseline, most participants preferred active (score 1-2, 37.2%) or collaborative (score 3, 44.9%) roles. The CASA intervention did not influence preferred decision-making roles (P > 0.1). Preferences significantly changed over 1 year (P < 0.01), becoming more active (82.1%, 84.2%, 89.0%, 90.1% active/collaborative at each timepoint). Among all models and covariates, there were no significant predictors of change (P > 0.1). Patients' preferred roles in decision making change over time, but changes are not well predicted. Clinicians should frequently and directly communicate with patients about their preferred decision-making roles.

Generalization of the relationships between sagittal balance geometric parameters

Sagittal balance of the spine characterizes its physiological alignment by muscular forces in the most effective way. Optimal spinal-pelvic relationships form a biomechanically efficient and optimal sagittal profile of the spine, which reduces energy consumption for an equilibrium state maintaining and also decrease the risk of degenerative damage to the intervertebral discs. Sagittal balance is described using the main angular parameters: pelvic incidence, lumbar lordosis, pelvic tilt and sacral slope. To form optimal spinal-pelvic relationships during surgical treatment surgeons measure the actual balance parameters of the patient and use empirical formulas to calculate their theoretical values at preoperative planning stage. To calculate the optimal parameters of the sagittal balance of a particular patient, various scientific groups have developed a number of empirical dependencies. However, there is no consensus in the scientific community on what specific formulas to use when planning treatment. This paper attempts to generalize the data available in the scientific literature on the parameters of the sagittal balance and to develop new generalizing regression dependencies that allow one to link the parameters of the sagittal balance. For this purpose, the source data used for constructing such dependencies were collected from the literature. The results of the sagittal balance measurement of Saratov State Medical University patients were also used in this study. We selected the initial numerical data linking the main parameters of the sagittal balance from the literature. The data was digitized and processed using statistics methods. Regression dependencies were obtained that allow calculating lumbar lordosis pelvic tilt, and sacral slope through the pelvic incidence of a particular patient. Generalizing dependences make it possible to systematize the clinical experience accumulated in the literature and were used in the development of the world's first system for preoperative planning of surgical treatment of the consequences and diseases of the spine-pelvic complex with biomechanical support of the doctor's decision.

Decision Support Collaborative Platform for e-Health Integration in Smart Communities Context

Data gathering related to biological parameters evolution using sensor networks associated to the patient, correlated with the analysis and data processing, together with a priori information and knowledge related with the patient, will allow developing new digital models for the characterization of healthcare state. By the help of algorithms for automated learning or neuromorphic devices, one can build a library for patients’ Digital Twins based on its corresponding health state.In this papers authors present the main components of an integrated informatic platform which assure an efficient access for patients to various medical health services. Also, the paper presents a model for each patient health state using specific artificial intelligence techniques as a support system for every family doctor's decision and medical experts.

Undressing the choice of dressings.

There is a vast number of wound dressings to choose from, which can make selecting the most suitable dressing for a wound a daunting prospect. This article highlights how the wound can guide the doctor's decision. Although all wound dressings are intended to achieve optimal wound healing and protection from contamination, they do so to varying degrees depending on the qualities of the dressing and the wound itself. This article looks at a range of dressings, from traditional to newer dressings, and evaluates their benefits and contraindications to help determine where they are best suited for use in wound management.

Thrombolytic Therapy in Treatment in Patients with Pulmonary Embolism not High-risk: SIRENA Registry Data

Aim. To study the features of the use of thrombolytic therapy (TLT) in normotensive patients with pulmonary embolism (PE) in real clinical practice in Russian hospitals.Material and Methods. From 04/1 5/2018 to 04/15/2019 patients hospitalized with a diagnosis of PE consistently were included in the Russian multicenter observational prospective register "SIRENA” (RusSIan REgistry of pulmoNAry embolism).Results. For 12 months in the registry was included 609 patients with a lifetime confirmed diagnosis of PE. TLT was performed in 152 patients with PE (25.0%), of which only 51 (33.8%) were indicated as "high risk" (shock or hypotension). In 101 not high risk patients, the indications for TLT were: severe shortness of breath/respiratory failure - 19 (18.8%), massive venous thrombosis - 7 (6.9%), signs of massive/submassive PE - 10 (9.9%), intermediate-high risk - 14 (13.9%), suspicion of acute coronary syndrome with ST segment elevation - 3 (2.9%), high pulmonary hypertension -2 (2.0%). The other 46 (45.5%) non-high-risk patients had no clear indication of the reasons for TLT in their medical history. To study the features of management of patients with not high-risk PE who received TLT (group 1), a selection of pairs of patients from the "SIRENA” registry, comparable in gender and age, in a ratio of 1:1 of patients with not high-risk PE who did not perform TLT (group 2). Hospital mortality was 4 (4%) patients in the TLT group and 6 (5.9%) patients in group 2 (р=0,748). Logistic regression analysis showed that floating blood clot in the veins of the lower extremities, syncopes in the debut of PE, respiratory rate over 22 per minute were independent clinical factors that significantly influence the doctor's decision to perform thrombolysis, and probability of completion TLT decreased in the presence of a history of bleeding, chronic kidney disease, surgery in the previous 12 months, increase in the size of the right atrium on EchoCG (statistical significance of the model x2=51.574; p&lt;0.001). The development of bleeding during hospitalization was recorded only in 10 (9.9%) patients of group 1, including severe (3 stage on the BARC scale) in 2 patients. Patients without TLT more often developed an acute heart failure (25.9% vs. 8.5%, p=0.043).Conclusion. In real clinical practice, there is a high frequency of TLT in patients with not high-risk PE. Floating blood clot in the veins of the lower extremities, syncope in the debut of PE, respiratory rate over 22 per minute were independent clinical factors that significantly influence the doctor's decision to perform thrombolysis.

Automatic identification of sweet spots from MERs for electrodes implantation in STN-DBS.

Microelectrode recordings (MERs) are a significant clinical indicator for sweet spots identification of implanted electrodes during deep brain stimulation of the subthalamic nucleus (STN) surgery. As 1D MERs signals have the unboundedness, large-range, large-amount and time-dependent characteristics, the purpose of this study is to propose an automatic and precise identification method of sweet spots from MERs, reducing the time-consuming and labor-intensive human annotations. We propose an automatic identification method of sweet spots from MERs for electrodes implantation in STN-DBS. To better imitate the surgeons' observation and obtain more intuitive contextual information, we first employ the 2D Gramian angular summation field (GASF) images generated from MERs data to perform the sweet spots determination for electrodes implantation. Then, we introduce the convolutional block attention module into convolutional neural network (CNN) to identify the 2D GASF images of sweet spots for electrodes implantation. Experimental results illustrate that the identification result of our method is consistent with the result of doctor's decision, while our method can achieve the accuracy and precision of 96.72% and 98.97%, respectively, which outperforms state-of-the-art for intraoperative sweet spots determination. The proposed method is the first time to automatically and accurately identify sweet spots from MERs for electrodes implantation by the combination an advanced time series-to-image encoding way with CBAM-enhanced networks model. Our method can assist neurosurgeons in automatically detecting the most likely locations of sweet spots for electrodes implantation, which can provide an important indicator for target selection while it reduces the localization error of the target during STN-DBS surgery.