Future prospects for partially ionized solar plasmas: the prominence case

Partially ionized plasmas (PIP) constitute an essential ingredient of our plasma universe. Historically, the physical effects associated with partial ionization were considered in astrophysical topics such as the interstellar medium, molecular clouds, accretion disks and, later on, in solar physics. PIP can be found in layers of the Sun's atmosphere as well as in solar structures embedded within it. As a consequence, the dynamical behaviour of these layers and structures is influenced by the different physical effects introduced by partial ionization. Here, rather than considering an exhaustive discussion of partially ionized effects in the different layers and structures of the solar atmosphere, we focus on solar prominences. The reason is that they represent a paradigmatic case of a partially ionized solar plasma, confined and insulated by the magnetic field, constituting an ideal environment to study the effects induced by partial ionization. We present the current knowledge about the effects of partial ionization in the global stability, mass cycle and dynamics of solar prominences. We revise the identified observational signatures of partial ionization in prominences. We conclude with prospects for PIP research in prominences, proposing the path for advancing in the prominence modelling and theory and using new and upcoming instrumentation. This article is part of the theme issue 'Partially ionized plasma of the solar atmosphere: recent advances and future pathways'.

Enhanced Nuclei Segmentation and Classification via Category Descriptors in the SAM Model

Segmenting and classifying nuclei in H&E histopathology images is often limited by the long-tailed distribution of nuclei types. However, the strong generalization ability of image segmentation foundation models like the Segment Anything Model (SAM) can help improve the detection quality of rare types of nuclei. In this work, we introduce category descriptors to perform nuclei segmentation and classification by prompting the SAM model. We close the domain gap between histopathology and natural scene images by aligning features in low-level space while preserving the high-level representations of SAM. We performed extensive experiments on the Lizard dataset, validating the ability of our model to perform automatic nuclei segmentation and classification, especially for rare nuclei types, where achieved a significant detection improvement in the F1 score of up to 12%. Our model also maintains compatibility with manual point prompts for interactive refinement during inference without requiring any additional training.

Attention guided multi-scale cluster refinement with extended field of view for amodal nuclei segmentation

Nuclei segmentation plays a crucial role in disease understanding and diagnosis. In whole slide images, cell nuclei often appear overlapping and densely packed with ambiguous boundaries due to the underlying 3D structure of histopathology samples. Instance segmentation via deep neural networks with object clustering is able to detect individual segments in crowded nuclei but suffers from a limited field of view, and does not support amodal segmentation. In this work, we introduce a dense feature pyramid network with a feature mixing module to increase the field of view of the segmentation model while keeping pixel-level details. We also improve the model output quality by adding a multi-scale self-attention guided refinement module that sequentially adjusts predictions as resolution increases. Finally, we enable clusters to share pixels by separating the instance clustering objective function from other pixel-related tasks, and introduce supervision to occluded areas to guide the learning process. For evaluation of amodal nuclear segmentation, we also update prior metrics used in common modal segmentation to allow the evaluation of overlapping masks and mitigate over-penalization issues via a novel unique matching algorithm. Our experiments demonstrate consistent performance across multiple datasets with significantly improved segmentation quality.

Mancha3D Code: Multipurpose Advanced Nonideal MHD Code for High-Resolution Simulations in Astrophysics

The Mancha3D code is a versatile tool for numerical simulations of magnetohydrodynamic (MHD) processes in solar/stellar atmospheres. The code includes nonideal physics derived from plasma partial ionization, a realistic equation of state and radiative transfer, which allows performing high-quality realistic simulations of magnetoconvection, as well as idealized simulations of particular processes, such as wave propagation, instabilities or energetic events. The paper summarizes the equations and methods used in the Mancha3D (Multifluid (-purpose -physics -dimensional) Advanced Non-ideal MHD Code for High resolution simulations in Astrophysics 3D) code. It also describes its numerical stability and parallel performance and efficiency. The code is based on a finite difference discretization and a memory-saving Runge-Kutta (RK) scheme. It handles nonideal effects through super-time-stepping and Hall diffusion schemes, and takes into account thermal conduction by solving an additional hyperbolic equation for the heat flux. The code is easily configurable to perform different kinds of simulations. Several examples of the code usage are given. It is demonstrated that splitting variables into equilibrium and perturbation parts is essential for simulations of wave propagation in a static background. A perfectly matched layer (PML) boundary condition built into the code greatly facilitates a nonreflective open boundary implementation. Spatial filtering is an important numerical remedy to eliminate grid-size perturbations enhancing the code stability. Parallel performance analysis reveals that the code is strongly memory bound, which is a natural consequence of the numerical techniques used, such as split variables and PML boundary conditions. Both strong and weak scalings show adequate performance up to several thousands of processors (CPUs).

Content preserving image translation with texture co-occurrence and spatial self-similarity for texture debiasing and domain adaptation

Models trained on datasets with texture bias usually perform poorly on out-of-distribution samples since biased representations are embedded into the model. Recently, various image translation and debiasing methods have attempted to disentangle texture biased representations for downstream tasks, but accurately discarding biased features without altering other relevant information is still challenging. In this paper, we propose a novel framework that leverages image translation to generate additional training images using the content of a source image and the texture of a target image with a different bias property to explicitly mitigate texture bias when training a model on a target task. Our model ensures texture similarity between the target and generated images via a texture co-occurrence loss while preserving content details from source images with a spatial self-similarity loss. Both the generated and original training images are combined to train improved classification or segmentation models robust to inconsistent texture bias. Evaluation on five classification- and two segmentation-datasets with known texture biases demonstrates the utility of our method, and reports significant improvements over recent state-of-the-art methods in all cases.

Epidemiological Characteristics of Deaths from COVID-19 in Peru during the Initial Pandemic Response

BACKGROUND AND AIM

Peru is the country with the highest mortality rate from COVID-19 globally, so the analysis of the characteristics of deaths is of national and international interest. The aim was to determine the epidemiological characteristics of deaths from COVID-19 in Peru from 28 March to 21 May 2020.

METHODS

Deaths from various sources were investigated, including the COVID-19 Epidemiological Surveillance and the National System of Deaths (SINADEF). In all, 3851 deaths that met the definition of a confirmed case and had a positive result of RT-PCR or rapid test IgM/IgG, were considered for the analysis. We obtained the epidemiological variables and carried out an analysis of time defined as the pre-hospital time from the onset of symptoms to hospitalization, and hospital time from the date of hospitalization to death.

RESULTS

Deaths were more frequent in males (72.0%), seniors (68.8%) and residents of the region of Lima (42.7%). In 17.8% of cases, the death occurred out-of-hospital, and 31.4% had some comorbidity. The median of pre-hospital time was 7 days (IQR: 4.0-9.0) and for the hospital time was 5 days (IQR: 3.0-9.0). The multivariable analysis with Poisson regression with robust variance found that the age group, comorbidity diagnosis and the region of origin significantly influenced pre-hospital time; while sex, comorbidity diagnosis, healthcare provider and the region of origin significantly influenced hospital time.

CONCLUSION

Deaths occurred mainly in males, seniors and on the coast, with considerable out-of-hospital deaths. Pre-hospital time was affected by age group, the diagnosis of comorbidities and the region of origin; while, hospital time was influenced by gender, the diagnosis of comorbidities, healthcare provider and the region of origin.

Methylene Blue functionalized carbon nanodots combined with different shape gold nanostructures for sensitive and selective SARS-CoV-2 sensing

The development of DNA-sensing platforms based on new synthetized Methylene Blue functionalized carbon nanodots combined with different shape gold nanostructures (AuNs), as a new pathway to develop a selective and sensitive methodology for SARS-CoV-2 detection is presented. A mixture of gold nanoparticles and gold nanotriangles have been synthetized to modify disposable electrodes that act as an enhanced nanostructured electrochemical surface for DNA probe immobilization. On the other hand, modified carbon nanodots prepared a la carte to contain Methylene Blue (MB-CDs) are used as electrochemical indicators of the hybridization event. These MB-CDs, due to their structure, are able to interact differently with double and single-stranded DNA molecules. Based on this strategy, target sequences of the SARS-CoV-2 virus have been detected in a straightforward way and rapidly with a detection limit of 2.00 aM. Moreover, this platform allows the detection of the SARS-CoV-2 sequence in the presence of other viruses, and also a single nucleotide polymorphism (SNPs). The developed approach has been tested directly on RNA obtained from nasopharyngeal samples from COVID-19 patients, avoiding any amplification process. The results agree well with those obtained by RT-qPCR or reverse transcription quantitative polymerase chain reaction technique.

P-366 Tumor necrosis factor-alpha and milk fat globule-epidermal growth factor 8: novel biomarkers to predict implantation failure and pregnancy loss

Study question

Can implantation failure (IF) and pregnancy loss be predicted in serum prior to in vitro fertilization (IVF)?

Summary answer

Tumor necrosis factor (TNFα) and milk fat globule-epidermal growth factor 8 (MFG-E8) levels may serve as serum markers to predict IF and pregnancy loss.

What is known already

In a normal pregnancy, mediators such as TNFα are released creating a physiological inflammatory response. However, an exaggerated release of TNFα has been associated with IF and recurrent pregnancy loss (RPL). Recent studies demonstrated that TNFα up-regulates the expression of inflammatory factors such as MFG-E8. MFG-E8 is known to modulate implantation by acting at various levels of the trophoblast and endometrial compartments. Hence an overexpression of this protein may result in apoptosis, endometrial damage, and impaired implantation.

Study design, size, duration

This multicentric prospective controlled pilot clinical study was conducted from January 2016 to January 2020 and included 30 women in their natural cycle in which serum MFG-E8, TNFα, estradiol (E2), and progesterone (P4) levels were quantified in the early proliferative (cycle day 2) and secretory phases (urinary LH + 7 days). Additionally, an endometrial biopsy was obtained on urinary LH + 7 days for MFG-E8 and TNF α protein and gene expression analysis.

Participants/materials, setting, methods

Women ages 21-35y were recruited from 3 groups: fertile controls (C), unexplained IF (following 3 failed good quality embryo transfers), and RPL (at least 2 unexplained first trimester miscarriages). Patients with history of uterine surgery, abnormal uterine cavity (fibroids, endometrial polyps, adhesions, adenomyosis, and congenital uterine abnormalities), hydrosalpinx, diminished ovarian reserve, harboring chromosomal rearrangements, thrombophilia, or autoimmune diseases were excluded.

Main results and the role of chance

Ten women were included in each group. No statistical differences were found in age, BMI, AMH, baseline FSH, and baseline antral follicle count among cohorts. Mean serum E2 and P4 levels were similar among all groups in both the proliferative and secretory phases: E2 proliferative (C 69.19±26.64  pg/ml, IF 64.19±32.56  pg/ml, RPL proliferative 57.44±38.51; p = 0.55), E2 secretory (C 164.10±52.57  pg/ml, IF 172.57±121, RPL 173.81±.97.35; p = 0.25), P4 proliferative (C 0.45±0.15 ng/ml, IF 0.45±0.19 ng/ml, RPL 0.53±0.18 ng/ml; p = 0.85), P4 secretory (C 7.42±4.06 ng/ml, IF 7.8±4.56 ng/ml, RPL 8.05±4.38 ng/ml; p = 0.74). Mean serum TNFα levels were significantly higher in both, the proliferative and secretory phases for the RPL group (proliferative RPL 9.98±4.47  pg/ml, IF 4.73±2.56  pg/ml, C 3.42±1.01  pg/ml; p = 0.001 vs secretory RPL 8.67±4.45  pg/ml, C 3.35±0.94  pg/ml, IF 3.85±1.01  pg/ml; p = 0.03). Mean serum MFG-E8 levels were significantly higher in the IF group during the proliferative phase (IF 373±201  pg/ml, RPL 201±115  pg/ml, C 225.58±109.73pg/ml; p = 0.03), but not in the secretory phase (IF 237±101  pg/ml, RPL 189±116  pg/ml, C 199.41±112.43  pg/ml; p = 0.15). Endometrial MFG-E8 mRNA levels were significantly lower in the IF and RPL group compared to C (p = 0.03). TNFα mRNA levels were not statistically significant among groups (p = 0.12).

Limitations, reasons for caution

This is a pilot study to assess feasibility. Due to the small sample size, the effects of more subtle covariates would not have been detected. Future larger studies are warranted.

Wider implications of the findings

These novels differentially expressed serum and endometrial markers may provide information on the physiology of implantation and could generate the basis for non-invasive diagnostic tools and therapeutic use of MFG-E8/TNFα antagonists in women with IF and RPL.

Trial registration number

IIT-2014-100366

P-704 Post trigger progesterone levels as a predictor of oocyte recovery rate

Study question

Do progesterone (P) levels the day after trigger for oocyte maturation impact the oocyte recovery rate?

Summary answer

A progesterone cutoff value of ≤ 5.0 ng/ml the day after trigger for oocyte maturation is associated with a lower oocyte recovery rate.

What is known already

Oocyte maturation and embryo development are controlled by intra-ovarian factors such as steroid hormones. P exists in the follicular fluid, and it is known to mediate some luteinizing hormone (LH)- initiated periovulatory events through autocrine/paracrine mechanisms that help mediate granulasa cell luteinization and oocyte maturation. More importantly, a rise in P levels is associated with an adequate follicular rupture. To date, no study has evaluated if lower P levels after oocyte maturation induction are a reflection of impaired physiological periovulatory mechanisms required for the oocyte release from the follicular wall.

Study design, size, duration

This monocentric retrospective analysis from January 2017 to December 2021 included 435 controlled ovarian hyperstimulation cycles for in vitro fertilization (IVF)/ egg freezing. Serum P, estradiol (E2), bHCG, and LH levels were measured the day after trigger for oocyte maturation to assure adequate luteinization.

Participants/materials, setting, methods

Women <4O years, that underwent controlled ovarian hyperstimulation with a GnRH antagonist protocol and final oocyte maturation (when ≥2 follicles reached ≥18 mm in diameter) induction with dual trigger (Leuprolide acetate and hCG) were segregated into two groups: Group 1: progesterone cutoff value of ≤ 5.0 ng/ml the day after trigger; Group 2: progesterone cutoff value of ≥ 5.1 ng/ml the day after trigger.

Main results and the role of chance

147 cycles in Group 1 were compared with 288 cycles on Group 2. No significant differences were noted in mean patient’s age, BMI, baseline FSH, AMH, baseline antral follicle count, serum P and E2 levels the day of trigger, total dose of gonadotropins, day of trigger as well as serum E2, LH, and bHCG levels the day after trigger among cohorts. A total of 5765 oocytes were retrieved: 1703 corresponding to group 1 and 4692 to group B. Women with progesterone levels higher than ≥5.1 ng/ml the day after trigger had a greater number of oocytes retrieved compared to women with a progesterone cutoff value of ≤ 5.0 ng/ml (16.3 ± 9.1 versus 11.5 ± 7.4, p = < 0.0001). The oocyte/follicle rate was also significantly higher in women in Group 2 vs Group 1 (80.3% vs 63.7%, p = 0.003)  However, the mature oocyte (MII) rate was comparable among cohorts (Group 1: 78.0 ± 20.2% vs Group 2: 78.8 ± 17.7%, p = 0.38).

Limitations, reasons for caution

The retrospective nature of the study, small sample size, selected progesterone cutoff value, and progesterone assay techniques compared to other ART centers may limit the external validity of our findings.

Wider implications of the findings

Our findings suggest for the first time that lower progesterone levels after final oocyte maturation induction may be a result of compromised mechanisms associated with the release of the oocyte from the follicular wall as demonstrated by a lower oocyte recovery rate.

Trial registration number

NA

Analysis of telephone demand in a heart failure unit: something more than a simple call

Introduction

Specific care programs for HF patients significantly reduce readmissions, which is why they are recommended in clinical practice guidelines with the highest level of evidence. Telephone assistance is a service offered to patients within these programs.

Purpose

The main objective was to evaluate the most frequent reasons for consultation and the ability to resolve them after the call.

Methods

This is a descriptive, retrospective study of all the telephone consultations registered between June 2020 and January 2021. The calls are attended by the HF consultation nurse, commenting with Cardiology on those that imply changes in the pharmacological treatment.

Sociodemographic and clinical variables, reasons for the call and their frequency were collected. Categorical variables were expressed as percentages and quantitative variables as means ± standard deviation.

Results

During the analyzed period, 1.608 consultations were attended, 494 were telephone calls, which represents 31% of the overall activity of the nursing consultation. 423 consultations were analyzed, 55.7% (235) reactive, and 44.3% (187) proactive. 72.5% of the patients were under follow-up by the nurse, compared to 27.5% who had completed drug titration.

The mean age was 65 (± 12.3) years, 65.6% were men, 98.4% had a caregiver, 77% of the patients had an LVEF <40%, Charlson modified mean 4.7 (± 3.6) and 92.3% were on treatment with beta-blockers, 89.1% with ACE inhibitors/ARBs/ARNI and 82.3% with MRA. The mean number of drugs per patient was 8.6 (± 3).

Reasons for the reactive calls were in 46% of the cases the consultation of symptoms, 20% to consult doubts about the treatment or procedures, 16% for problems related to the prescription of pharmacological treatment and 14% for the appointment management.

Reasons for proactive calls were 40% for symptoms control, 30.5% for appointment management, 27.3% for drug titration and 2.1% related to other causes.

When the reason for consultation was the appearance of symptoms, the mean number of calls required was 1.4 (± 0.8), 77.6% of the calls were solved through the telephone, 16.4% required a face-to-face visit in the nursing consultation and 2.7% were referred to the emergency room. Queries related to doubts, problems with the prescription or appointment management were solved in a single telephone consultation.

Conclusions

The most common telephone request when the patient made the call was the worsening of HF symptoms. The telephone consultation is presented as an effective tool in these cases, by favoring the accessibility of patients and avoiding unnecessary trips to the health center.

Funding Acknowledgement

Type of funding sources: None. Table 1. Descriptive details of patientsTable 2. Call reasons

Analysis of LpA levels in young patients after ACS: getting to know less famous risk factors

Introduction

Lipoprotein A (LpA) has been shown to be an emerging risk factor, proposing that values greater than 60 mg/dl increases cardiovascular risk. There are few data about LpA values in young patients who have suffered a major cardiovascular event.

Purpose

The objective of this work was to describe the LpA values observed in young patients admitted for acute coronary syndrome in our center, and subsequently to compare these values according to the patients' previous cardiovascular risk.

Methods

This is a descriptive and observational study, in which all male patients under 65 years and women under 70 years who have suffered STEMI or NSTEMI from November 2019 to February 2021 admitted to our center were consecutively included. In addition to LpA values, the following variables were recollected: age, sex, high blood pressure, diabetes mellitus, dyslipidemia, stroke, chronic kidney injury, smoking, alcoholism, toxics, total cholesterol and SCORE risk.

Results

159 patients were included. The mean of LpA value was 41,08 mg/dl (standard deviation 38, range 1–155, percentile 25th: 9,7; percentile 50th: 28,8; percentile 75th: 59,1). 24,5% presented levels of LpA greater than 60 mg/dl. The percentage of patients with LpA levels >60 mg was 32,4% in low SCORE group and 22,4% in greater than low SCORE group without significant differences.

The table compares the LpA values according to the cardiovascular risk SCORE those patients presented before the acute coronary syndrome (low SCORE vs moderate, high or very high SCORE). As we can see in the table, we found a trend to present higher LpA values in patients with low SCORE risk compared to those with higher than low SCORE risk, without reaching statistical significance.

Conclusions

In a sample of young patients with acute coronary syndrome, the LpA mean was 41,08 mg/dl. 24,5% of patients had values of LpA greater than 60 mg/dl. No significant differences were found according to the SCORE prior to the event, although there was a non-significant trend towards a higher LpA in patients with low SCORE.

Funding Acknowledgement

Type of funding sources: None. Table 1. LpA values

P–685 Luteinization after final oocyte maturation induction is not compromised in women that receive double dose of Gonadotrophin-releasing hormone antagonists during controlled ovarian hyperstimulation

Study question

Does the use of double dose of GnRH antagonists during COH in women with risk of premature LH surge alter luteinization after final oocyte maturation induction?

Summary answer

The use of double dose of GnRH antagonist in women with risk of premature luteinizing hormone surge dosent affect luteinization after final oocyte maturation induction.

What is known already

GnRH antagonists are used to prevent a premature LH surge during controlled ovarian hyperstimulation. The antagonists directly inhibit gonadotrophin release within several hours through competitive binding to pituitary GnRH receptors, producing a rapid suppression of LH and FSH, with no initial flare effect. In women with diminished ovarian reserve (DOR) it is not uncommon that premature luteinization cannot be completely prevented using a daily dose GnRH antagonist. To date, no study has evaluated the effects of using a daily double dose of GnRH antagonists to prevent a premature LH surge and its effect on luteinization after final oocyte maturation induction.

Study design, size, duration

This monocentric retrospective analysis evaluated the effect on luteinization after final oocyte maturation induction in twenty women during COH who received a daily double dose of GnRH antagonists (Cetrotide 0.25 mg/mL, Merck) from January 2020 to December 2020.

Participants/materials, setting, methods

Women with severe DOR and history of premature luteinization during COH received a double dose of GnRH antagonist when the leading follicle reached 12–14 mm (am and pm). When two follicles reached ≥18 mm in diameter, final oocyte maturation was induced with dual trigger using Leuprolide acetate and hCG. Progesterone, estradiol, bHCG, and LH levels were measured the day after final oocyte maturation induction to assure adequate luteinization.

Main results and the role of chance

In total twenty women were included in the analysis. Mean age 36.8± 4.2, AMH 0.65± 0.32 ng/ml, baseline antral follicle count 4± 2.3, serum hormone levels the day of ovulation induction trigger: progesterone 0.89± 0.34 ng/ml, LH 1.6± 2.1 ng/ml, estradiol 1235 ± 1420 pg/ml. Post-surge serum hormone levels average reached adequate levels: estradiol 1645 ± 1116 pg/ml, progesterone 20.4 ±2.2 ng/ml, LH 62.66± 10.5 IU/ml and, bHCG 247±115 IU/ml. A total of 76 oocytes were retrieved (3.8± 0.8 oocytes per patient), 63.1% (48/76) MII, 22% (17/76) MI, 14% (11/76) GV.

Limitations, reasons for caution

The retrospective nature of the study, small sample size, and potential variability in the study center’s laboratory protocol(s) compared to other reproductive treatment centers may limit the external validity of our findings.

Wider implications of the findings: The daily use of double dose of GnRH antagonists during COH offers the possibility of preventing a premature LH surge in women with DOR with high risk of early ovulation, without compromising luteinization after final oocyte maturation induction.

Trial registration number

NA

Dual attention multiple instance learning with unsupervised complementary loss for COVID-19 screening

Chest computed tomography (CT) based analysis and diagnosis of the Coronavirus Disease 2019 (COVID-19) plays a key role in combating the outbreak of the pandemic that has rapidly spread worldwide. To date, the disease has infected more than 18 million people with over 690k deaths reported. Reverse transcription polymerase chain reaction (RT-PCR) is the current gold standard for clinical diagnosis but may produce false positives; thus, chest CT based diagnosis is considered more viable. However, accurate screening is challenging due to the difficulty in annotation of infected areas, curation of large datasets, and the slight discrepancies between COVID-19 and other viral pneumonia. In this study, we propose an attention-based end-to-end weakly supervised framework for the rapid diagnosis of COVID-19 and bacterial pneumonia based on multiple instance learning (MIL). We further incorporate unsupervised contrastive learning for improved accuracy with attention applied both in spatial and latent contexts, herein we propose Dual Attention Contrastive based MIL (DA-CMIL). DA-CMIL takes as input several patient CT slices (considered as bag of instances) and outputs a single label. Attention based pooling is applied to implicitly select key slices in the latent space, whereas spatial attention learns slice spatial context for interpretable diagnosis. A contrastive loss is applied at the instance level to encode similarity of features from the same patient against representative pooled patient features. Empirical results show that our algorithm achieves an overall accuracy of 98.6% and an AUC of 98.4%. Moreover, ablation studies show the benefit of contrastive learning with MIL.

Quantitative Assessment of Chest CT Patterns in COVID-19 and Bacterial Pneumonia Patients: a Deep Learning Perspective

BACKGROUND

It is difficult to distinguish subtle differences shown in computed tomography (CT) images of coronavirus disease 2019 (COVID-19) and bacterial pneumonia patients, which often leads to an inaccurate diagnosis. It is desirable to design and evaluate interpretable feature extraction techniques to describe the patient's condition.

METHODS

This is a retrospective cohort study of 170 confirmed patients with COVID-19 or bacterial pneumonia acquired at Yeungnam University Hospital in Daegu, Korea. The Lung and lesion regions were segmented to crop the lesion into 2D patches to train a classifier model that could differentiate between COVID-19 and bacterial pneumonia. The K-means algorithm was used to cluster deep features extracted by the trained model into 20 groups. Each lesion patch cluster was described by a characteristic imaging term for comparison. For each CT image containing multiple lesions, a histogram of lesion types was constructed using the cluster information. Finally, a Support Vector Machine classifier was trained with the histogram and radiomics features to distinguish diseases and severity.

RESULTS

The 20 clusters constructed from 170 patients were reviewed based on common radiographic appearance types. Two clusters showed typical findings of COVID-19, with two other clusters showing typical findings related to bacterial pneumonia. Notably, there is one cluster that showed bilateral diffuse ground-glass opacities (GGOs) in the central and peripheral lungs and was considered to be a key factor for severity classification. The proposed method achieved an accuracy of 91.2% for classifying COVID-19 and bacterial pneumonia patients with 95% reported for severity classification. The CT quantitative parameters represented by the values of cluster 8 were correlated with existing laboratory data and clinical parameters.

CONCLUSION

Deep chest CT analysis with constructed lesion clusters revealed well-known COVID-19 CT manifestations comparable to manual CT analysis. The constructed histogram features improved accuracy for both diseases and severity classification, and showed correlations with laboratory data and clinical parameters. The constructed histogram features can provide guidance for improved analysis and treatment of COVID-19.

When Is Sentinel Node Biopsy Indicated in High-Risk Ductal Carcinoma in situ? Four Hundred Sixty-Eight Cases from Three Institutions

<b><i>Introduction:</i></b> Sentinel lymph node biopsy (SLNB) in ductal carcinoma in situ (DCIS) is not indicated. However, in certain cases (size &#x3e;3 cm, high grade, mass effect on mammography, or palpable mass), it may be possible to find incidental invasive carcinoma (IC) that requires an SLNB. We studied the correlation of the aforesaid factors with the probability of finding IC in the surgical specimen. <b><i>Methods:</i></b> Data was collected from 3 different institutions between 2010 and 2016, recording characteristics such as, but not limited to: high grade, size &#x3e;3 cm, mass effect on mammography, and palpable mass. <b><i>Results:</i></b> On the whole, 468 “high-risk” DCIS cases were identified, 139 (29%) of which had IC. When the DCIS was high grade or the size was &#x3e;3 cm, there was no significant difference in the probability of finding IC in the surgical specimen (OR = 1.13; 95% CI 0.84–1.51; OR = 1.2; 95% CI 0.85–1.40). Nevertheless, when a high grade and size (&#x3e;3 cm) were combined, IC was more likely to exist (72.7 vs. 27.3%; <i>p</i> = 0.001). In addition, mass effect and palpation were independently associated with a significantly greater degree of IC (OR = 12.76; 95% CI 6.93–23.52). <b><i>Conclusions:</i></b> The results suggest that high-grade DCIS or DCIS with a size &#x3e;3 cm, independently, does not require SLNB. Nonetheless, in the event that both factors are found in the same case, SLNB may be indicated. Additionally, SLNB is advisable for DCIS cases that are palpable or show a mass effect on mammography.

Waddling beyond door to balloon times and impinging true ischemic times with artificial intelligence-guided single lead EKG for STEMI detection

Background

The present process of STEMI detection is cumbersome as it utilizes outdated equipment and requires a trained technician and an expert cardiologist. We have developed a patient-administered, Artificial Intelligence (AI) guided, Single Lead EKG for early STEMI detection.

Purpose

To answer the question “Is early STEMI detection possible with a Single Lead EKG?”

Methods

We experimented with an AI-guided algorithm for a single-lead EKG for STEMI detection with the following step-wise developments: 1) An AI algorithm that predictably interprets STEMI using a 12-lead EKG; 2) An AI algorithm for STEMI detection using a single-lead EKG; 3) A methodology for identifying the best single lead to detect STEMI; 4) Advanced AI algorithms for STEMI localization with a single-lead EKG. The AI methodology was as follows: Sample: The mammoth Latin American Telemedicine Infarct Network telemedicine database that provides an umbrella of AMI management to 100 million patients in Brazil, Colombia, Mexico, Chile, and Argentina was queried for cardiologist annotated EKG. A total of 8,511 EKG and 90,592 classified heartbeats were selected for the experiments. Preprocessing: segmentation of each ECG into individual heartbeats. Training & Testing: 90% and 10%, respectively, of the total dataset. Classification: 1-D Convolutional Neural Network; classes were construed for each heartbeat. Performance indicators were calculated per lead.

Results

The algorithm was able to provide an accuracy of 91.9%. Lead V2 yielded the best results among individual leads for STEMI detection.

Conclusions

Early experiments provide a framework for augmenting STEMI detection with the use of AI-guided, single lead techniques. Such approaches seem rational as we target the reduction of true STEMI ischemic times.

Funding Acknowledgement

Type of funding source: None

Moving in sync – concordance betweena artificial intelligence and cardiologist on detecting normal electrocardiograms

Background

Merging modern technologies with classic diagnostic tests often results in a sense of insecurity within the medical community, particularly so with potentially life-saving studies such as the electrocardiogram (EKG). In order to provide a greater sense of trust between Artificial Intelligence (AI) and cardiologists, we provide an AI-driven algorithm capable of accurately and reliably characterize an EKG as normal within a highly complex, cardiologist-reviewed EKG database and report the degree of concordance between this machine vs physician scenario.

Purpose

To provide a dependable and accurate AI algorithm that conducts EKG interpretation in a cardiologist-tier manner.

Methods

The International Telemedical System (ITMS) developed and tested an EKG assessing AI algorithm and incorporated it into the workflow of their Telemedicine Integrated Platform, a digital EKG reading program where cardiologists continuously report their findings remotely in real-time. During the month of April 2,019; 35 ITMS cardiologists reported a grand total of 61,441 EKG records, later submitting them to the AI algorithm implemented through the “One Click Report” process. Through this simple 2-step approach, the algorithm provides a suggestion of “Normal” or “Abnormal” to the cardiologist based on the patterns of the fiducial points included in said EKG reports. Confirmation of these suggestions by the cardiologists ensued.

Results

Overall, cardiologists confirmed 23,213 out of 25,013 AI outputs for “Normal” EKGs, demonstrating a concordance of 92.8% for Normal diagnosis.

Conclusion

Through this methodology, we provide an AI technology that can be reliably applied and trusted in EKG digital platforms to identify and suitably label a normal EKG. Further testing will accrue into a multi label algorithm compatible with abnormal cardiovascular entities, potentially precluding the role of the cardiologist for triaging, particularly in the prehospital setting. We anticipate that this approach will become a promising methodology in modern cardiology practice.

Funding Acknowledgement

Type of funding source: None

Baby steps in the path of modifying the role of cardiologists for interpreting EKG for AMI

Background

EKG interpretation is slowly transitioning to a physician-free, Artificial Intelligence (AI)-driven endeavor. Our continued efforts to innovate follow a carefully laid stepwise approach, as follows: 1) Create an AI algorithm that accurately identifies STEMI against non-STEMI using a 12-lead EKG; 2) Challenging said algorithm by including different EKG diagnosis to the previous experiment, and now 3) To further validate the accuracy and reliability of our algorithm while also improving performance in a prehospital and hospital settings.

Purpose

To provide an accurate, reliable, and cost-effective tool for STEMI detection with the potential to redirect human resources into other clinically relevant tasks and save the need for human resources.

Methods

Database: EKG records obtained from Latin America Telemedicine Infarct Network (Mexico, Colombia, Argentina, and Brazil) from April 2014 to December 2019. Dataset: A total of 11,567 12-lead EKG records of 10-seconds length with sampling frequency of 500 [Hz], including the following balanced classes: unconfirmed and angiographically confirmed STEMI, branch blocks, non-specific ST-T abnormalities, normal and abnormal (200+ CPT codes, excluding the ones included in other classes). The label of each record was manually checked by cardiologists to ensure precision (Ground truth). Pre-processing: The first and last 250 samples were discarded as they may contain a standardization pulse. An order 5 digital low pass filter with a 35 Hz cut-off was applied. For each record, the mean was subtracted to each individual lead. Classification: The determined classes were STEMI (STEMI in different locations of the myocardium – anterior, inferior and lateral); Not-STEMI (A combination of randomly sampled normal, branch blocks, non-specific ST-T abnormalities and abnormal records – 25% of each subclass). Training & Testing: A 1-D Convolutional Neural Network was trained and tested with a dataset proportion of 90/10; respectively. The last dense layer outputs a probability for each record of being STEMI or Not-STEMI. Additional testing was performed with a subset of the original dataset of angiographically confirmed STEMI.

Results

See Figure Attached – Preliminary STEMI Dataset Accuracy: 96.4%; Sensitivity: 95.3%; Specificity: 97.4% – Confirmed STEMI Dataset: Accuracy: 97.6%; Sensitivity: 98.1%; Specificity: 97.2%.

Conclusions

Our results remain consistent with our previous experience. By further increasing the amount and complexity of the data, the performance of the model improves. Future implementations of this technology in clinical settings look promising, not only in performing swift screening and diagnostic steps but also partaking in complex STEMI management triage.

Funding Acknowledgement

Type of funding source: None

Artificial intelligence methodology: multi-label classification of abnormal EKG records

Background

Our previous experience with Artificial Intelligence (AI)-conducted EKG characterization displayed outstanding results in fast and reliable identification of Normal EKGs within the International Telemedical System (ITMS)'s massive record repository. By expanding the array of recognizable cardiovascular entities, we upgraded our methodology to accurately discriminate an anomaly amongst a highly complex database of EKG records.

Purpose

To present a feasible AI-guided filter that can accurately discriminate and classify Normal and Abnormal EKG records within a multilabeled cardiologist-annotated EKG database.

Methods

ITMS developed and tested the “One Click”' process, a “Normal/Abnormal” EKG assessing AI algorithm, by incorporating it into their digital EKG reading platform where cardiologists continuously report their findings remotely in real time. To ameliorate the diagnostic range of the algorithm, a separate dataset of 121,641 12-lead EKG records was consolidated from the ITMS database from October 2011 to January 2019. Only de-identified data was used. Preprocessing: The first 2s of each short lead and 9s of the long lead were considered. Limb leads I, II and III; and precordial leads V1, V2, V3, and V5 were used. The mean was removed from each lead. AI models/Classification: Two models were created and tested independently based on the method of EKG acquisition (69,852 records transtelephonic [TTP]; 52,259 mobile transmission [MOB]). Each record is categorized into six disjoint classes based on the most common types of cardiac disorders (Low/null co-occurrence pathologies in these datasets were grouped into analogous groups). Training/Testing: Distribution of both sets per transmission type was performed through a greedy algorithm, which identified multiple diagnoses per EKG record and labeled it separately to the corresponding group, ensuring sufficient samples per class. Detailed class distribution is shown below. An inception convolutional neural network was implemented; “Normal” or “Abnormal” labels were assigned to each EKG record independently and were compared to cardiologists' reports; performance indicators were calculated for each model and group.

Results

MOB model accrued an average accuracy of 86.7%; sensitivity of 90.5%; and specificity of 83.9%. TTP model yielded an average accuracy of 77.2%; sensitivity of 91.1%; and specificity of 69.4% (Lower values were attributed to the “Ventricular Complexes” group, which challenged the algorithm by having a smaller ratio of abnormal exams). Detailed results of each training set are shown below.

Conclusion

Providing an effective and reliable multilabel-capable EKG triaging tool remains a challenging but attainable goal. Continuous systematic enhancement of our AI-driven methodology has led us to satisfactory, yet imperfect results which compel us to further study and improve our efforts to provide a trustworthy cardiologist-friendly triage device.

Funding Acknowledgement

Type of funding source: None

Standardized Assessment of Automatic Segmentation of White Matter Hyperintensities and Results of the WMH Segmentation Challenge

Quantification of cerebral white matter hyperintensities (WMH) of presumed vascular origin is of key importance in many neurological research studies. Currently, measurements are often still obtained from manual segmentations on brain MR images, which is a laborious procedure. The automatic WMH segmentation methods exist, but a standardized comparison of the performance of such methods is lacking. We organized a scientific challenge, in which developers could evaluate their methods on a standardized multi-center/-scanner image dataset, giving an objective comparison: the WMH Segmentation Challenge. Sixty T1 + FLAIR images from three MR scanners were released with the manual WMH segmentations for training. A test set of 110 images from five MR scanners was used for evaluation. The segmentation methods had to be containerized and submitted to the challenge organizers. Five evaluation metrics were used to rank the methods: 1) Dice similarity coefficient; 2) modified Hausdorff distance (95th percentile); 3) absolute log-transformed volume difference; 4) sensitivity for detecting individual lesions; and 5) F1-score for individual lesions. In addition, the methods were ranked on their inter-scanner robustness; 20 participants submitted their methods for evaluation. This paper provides a detailed analysis of the results. In brief, there is a cluster of four methods that rank significantly better than the other methods, with one clear winner. The inter-scanner robustness ranking shows that not all the methods generalize to unseen scanners. The challenge remains open for future submissions and provides a public platform for method evaluation.

Cutaneous Manifestations of Zika in Peru

This article aims to describe the cutaneous manifestations observed in the Zika epidemic in Peru during 2016 and 2017, as well as discuss the potential differential diagnoses. During the outbreak, the main reason for seeking medical advice was the development of a pruriginous maculopapular rash with a marked papular component, which started on the chest and later generalized to the rest of the body. Similar manifestations were noted in adults, children, and pregnant women. Other manifestations such as conjunctivitis, edema, or petechiae on the palate were rare. We suggest that in areas that are endemic for arboviral infections, in the differential diagnosis of a rash one must consider infections such as dengue, Zika, or chikungunya viruses. In nonendemic areas, the diagnosis is more difficult, as the rash may result from other viral infections not transmitted by arthropods and/or reactive or inflammatory diseases (urticaria, atopic dermatitis, systemic lupus erythematosus). We recommend that primary care health personnel are trained in the recognition of the mucocutaneous lesions caused by Zika virus infection, which could contribute to the identification of suspicious cases, particularly pregnant women.

3,5-T2 and 3,3',5-T3 Regulate Cerebellar Thyroid Hormone Signalling and Myelin Molecular Dynamics in Tilapia

In contrast to mammalian adults, myelination in teleosts occurs throughout their lifespan and most of the progenitor cells are originated in the cerebellum. To understand the role that thyroid hormones (THs) play in juvenile cerebellar myelination in teleosts, we identified and localised the expression of genes involved in TH signalling (mct8, oatp1c1, dio2, dio3, thraa and l-thrb1) and analysed the effects of the two bioactive THs, T2 and T3, upon their regulation, as well as upon some structural components of the myelination process. Ex vivo approaches using organotypic cerebellar cultures followed by FISH and qPCR showed gene-specific localisation and regulation of TH signalling genes in the cerebellar nuclei. In vivo approaches using methimazole (MMI)-treated juvenile tilapias replaced with low doses of T3 and T2 showed by immunofluorescence that myelin fibres in the cerebellum are more abundant in the granular layer and that their visible size is reduced after MMI treatment but partially restored with TH replacement, suggesting that low doses of TH promote the re-myelination process in an altered condition. Together, our data support the idea that T2 and T3 promote myelination via different pathways and prompt T2 as a target for further analysis as a promising therapy for hypomyelination.

Growth hormone protects against kainate excitotoxicity and induces BDNF and NT3 expression in chicken neuroretinal cells

There is increasing evidence to suggest a beneficial neuroprotective effect of growth hormone (GH) in the nervous system. While our previous studies have largely focused on retinal ganglion cells (RGCs), we have also found conclusive evidence of a pro-survival effect of GH in cells of the inner nuclear layer (INL) as well as a protective effect on the dendritic trees of the inner plexiform layer (IPL) in the retina. The administration of GH in primary neuroretinal cell cultures protected and induced neural outgrowths. Our results, both in vitro (embryo) and in vivo (postnatal), showed neuroprotective actions of GH against kainic acid (KA)-induced excitotoxicity in the chicken neuroretina. Intravitreal injections of GH restored brain derived neurotrophic factor (BDNF) expression in retinas treated with KA. In addition, we demonstrated that GH over-expression and exogenous administration increased BDNF and neurotrophin-3 (NT3) gene expression in embryonic neuroretinal cells. Thus, GH neuroprotective actions in neural tissues may be mediated by a complex cascade of neurotrophins and growth factors which have been classically related to damage prevention and neuroretinal tissue repair.

Verbal fluency tests: Normative data for the Latin American Spanish speaking adult population

OBJECTIVE

To generate normative data for the Verbal Fluency Tests across 11 countries in Latin America, with country-specific adjustments for gender, age, and education, where appropriate.

METHOD

The sample consisted of 3,977 healthy adults who were recruited from Argentina, Bolivia, Chile, Cuba, El Salvador, Guatemala, Honduras, Mexico, Paraguay, Peru, and, Puerto Rico. Each subject was administered the Verbal Fluency Test as part of a larger neuropsychological battery. A standardized five-step statistical procedure was used to generate the norms.

RESULTS

The final multiple linear regression models for the letter F explained 8-30% of the variance, 7-32% for letter A, 8-32% for the letter S, and 16-43% for the animal category in Verbal Fluency Test scores. Although t-tests showed significant differences between men and women on the Verbal Fluency Test, they did not have an effect size larger than 0.3. As a result, gender-adjusted norms were not generated.

CONCLUSIONS

This is the first normative multicenter study conducted in Latin America aiming to create norms for the Verbal Fluency Test; this study will have important outcomes for the future of neuropsychology in the region.

Hopkins Verbal Learning Test- Revised: Normative data for the Latin American Spanish speaking adult population

OBJECTIVE

To generate normative data on the Hopkins Verbal Learning Test- Revised (HVLT-R) across 11 countries in Latin America, with country-specific adjustments for gender, age, and education, where appropriate.

METHOD

The sample consisted of 3,977 healthy adults who were recruited from Argentina, Bolivia, Chile, Cuba, El Salvador, Guatemala, Honduras, Mexico, Paraguay, Peru, and, Puerto Rico. Each subject was administered the HVLT-R as part of a larger neuropsychological battery. A standardized five-step statistical procedure was used to generate the norms.

RESULTS

The final multiple linear regression models explained 17- 45% of the variance in HVLT-R scores. Although t-tests showed significant differences between men and women in Guatemala on the HVLT-R, it was a small effect size. As a result, gender-adjusted norms were not generated.

CONCLUSIONS

The results from this study will have a substantial impact on the practice of neuropsychology in Latin America, as this is the first normative multicenter study to develop norms for the HVLT-R in this region.

Rey-Osterrieth Complex Figure - copy and immediate recall: Normative data for the Latin American Spanish speaking adult population

OBJECTIVE

To generate normative data on the Rey-Osterrieth Complex Figure Test (ROCF) across 11 countries in Latin America, with country-specific adjustments for gender, age, and education, where appropriate.

METHOD

The sample consisted of 3,977 healthy adults who were recruited from Argentina, Bolivia, Chile, Cuba, El Salvador, Guatemala, Honduras, Mexico, Paraguay, Peru, and, Puerto Rico. Each subject was administered the ROCF as part of a larger neuropsychological battery. A standardized five-step statistical procedure was used to generate the norms.

RESULTS

The final multiple linear regression models explained 7-34% of the variance in ROCF copy scores and 21-41% of the variance in immediate recall scores. Although t-tests showed significant differences between men and women on ROCF copy and immediate recall scores, none of the countries had an effect size larger than 0.3. As a result, gender-adjusted norms were not generated.

CONCLUSIONS

The present study is the first to create norms for the ROCF in Latin America. As a result, this study will have important implications for the formation and practice of neuropsychology in this region.

[What is the true role of surgery in the infertility associated with endometriosis?]

BACKGROUND

Endometriosis is one of the most challenging clinical conditions for gynecologists. Associated pain and infertility are often difficult to manage, and current treatment strategies remain limited.

OBJECTIVE

This review reviews the current scientific evidence for general gynecologist and provides an overview of current information regarding the treatment of patients with endometriosis-related infertility, offering strong evidence to consider a less invasive approach, and highlights potential hazards of surgery within patients desiring to achieve a pregnancy.

Preparation of human resources for future nuclear energy using FBNR as the instrument of learning

An increasing number of developing countries are showing interest to become the emerging countries to nuclear energy. Most of these countries lack human resources and adequate infrastructures to enter such a venture. The principle objective of activities of FBNR Group is to train human resources for the countries that at the present lack the necessary conditions, but aim at the future clean and safe nuclear energy through the fourth generation and INPRO compatible nuclear reactors. The preparation for the future nuclear energy is done through development of innovative nuclear reactor that meets the INPRO philosophies and criteria. These countries may or may not have decided as yet to utilize nuclear energy, but are interested to gain a strong educational foundation for their future. The research and development of a small innovative nuclear reactor FBNR is used as the instrument for learning. The young scientists will learn how to be innovative with the vision of INPRO philosophy and criteria.

Pressure-controlled injection of guar gum stabilized microscale zerovalent iron for groundwater remediation

The paper reports a pilot injection test of microsized zerovalent iron (mZVI) dispersed in a guar gum shear thinning solution. The test was performed in the framework of the EU research project AQUAREHAB in a site in Belgium contaminated by chlorinated aliphatic hydrocarbons (CAHs). The field application was aimed to overcome those critical aspects which hinder mZVI field injection, mainly due to the colloidal instability of ZVI-based suspensions. The iron slurry properties (iron particles size and concentration, polymeric stabilizer type and concentration, slurry viscosity) were designed in the laboratory based on several tests (reactivity tests towards contaminants, sedimentation tests and rheological measurements). The particles were delivered into the aquifer through an injection well specifically designed for controlled-pressure delivery (approximately 10 bars). The well characteristics and the critical pressure of the aquifer (i.e. the injection pressure above which fracturing occurs) were assessed via two innovative injection step rate tests, one performed with water and the other one with guar gum. Based on laboratory and field preliminary tests, a flow regime at the threshold between permeation and preferential flow was selected for mZVI delivery, as a compromise between the desired homogeneous distribution of the mZVI around the injection point (ensured by permeation flow) and the fast and effective injection of the slurry (guaranteed by high discharge rates and injection pressure, resulting in the generation of preferential flow paths). A monitoring setup was designed and installed for the real-time monitoring of relevant parameters during injection, and for a fast determination of the spatial mZVI distribution after injection via non-invasive magnetic susceptibility measurements.

Modelling Ecological Cognitive Rehabilitation Therapies for Building Virtual Environments in Brain Injury

BACKGROUND

Brain Injury (BI) has become one of the most common causes of neurological disability in developed countries. Cognitive disorders result in a loss of independence and patients' quality of life. Cognitive rehabilitation aims to promote patients' skills to achieve their highest degree of personal autonomy. New technologies such as virtual reality or interactive video allow developing rehabilitation therapies based on reproducible Activities of Daily Living (ADLs), increasing the ecological validity of the therapy. However, the lack of frameworks to formalize and represent the definition of this kind of therapies can be a barrier for widespread use of interactive virtual environments in clinical routine.

OBJECTIVES

To provide neuropsychologists with a methodology and an instrument to design and evaluate cognitive rehabilitation therapeutic interventions strategies based on ADLs performed in interactive virtual environments.

METHODS

The proposed methodology is used to model therapeutic interventions during virtual ADLs considering cognitive deficit, expected abnormal interactions and therapeutic hypotheses. It allows identifying abnormal behavioural patterns and designing interventions strategies in order to achieve errorless-based rehabilitation.

RESULTS

An ADL case study ('buying bread') is defined according to the guidelines established by the ADL intervention model. This case study is developed, as a proof of principle, using interactive video technology and is used to assess the feasibility of the proposed methodology in the definition of therapeutic intervention procedures.

CONCLUSIONS

The proposed methodology provides neuropsychologists with an instrument to design and evaluate ADL-based therapeutic intervention strategies, attending to solve actual limitation of virtual scenarios, to be use for ecological rehabilitation of cognitive deficit in daily clinical practice. The developed case study proves the potential of the methodology to design therapeutic interventions strategies; however our current work is devoted to designing more experiments in order to present more evidence about its values.