Prognostic significance of the Holter-derived T-wave variability in patients with ventricular tachyarrhythmias complicating acute coronary syndrome-TWIST study.

BACKGROUND: We aimed to investigate the association between ventricular repolarization instability and sustained ventricular tachycardia and ventricular fibrillation (VT/VF) occurring within 48 h (acute-phase VT/VF) after the onset of acute coronary syndrome (ACS) and the prognostic role of repolarization instability and heart rate variability (HRV) after discharge from the hospital. METHODS: We studied 572 ACS patients with a left ventricular ejection fraction >35%. The ventricular repolarization instability was assessed by the beat-to-beat T-wave amplitude variability (TAV) using high-resolution 24-h Holter ECGs recorded at a median of 11 days from the date of admission. We calculated the HRV parameters including the deceleration capacity (DC) and non-Gaussian index calculated on a 25 s timescale (λ25s). The DC and λ25s were dichotomized based on previous studies' thresholds. RESULTS: Acute-phase VT/VF developed in 43 (7.5%) patients. In-hospital mortality was significantly higher among VT/VF patients (4.7% vs. 0.9%, p = .03). An adjusted logistic model showed that the maximum TAV (odds ratio 1.02, 95% confidence interval [CI] 1.00-1.29, p = .04) was associated with acute-phase VT/VF. During a median follow-up period of 2.1 years, 19 (3.3%) patients had cardiac deaths or resuscitated cardiac arrest. Acute-phase VT/VF (p = .12) and TAV (p = .72) were not significant predictors of survival. An age and sex-adjusted Cox model showed that the DC (p < .01), λ25s (p < .01), and emergency coronary intervention (p < .01) were independent predictors. CONCLUSION: T-wave amplitude variability was associated with acute-phase VT/VF, but the TAV was not predictive of survival post-discharge. The DC, λ25s, and emergency coronary intervention were independent predictors of survival.

Differences between human and machine perception in medical diagnosis.

Deep neural networks (DNNs) show promise in image-based medical diagnosis, but cannot be fully trusted since they can fail for reasons unrelated to underlying pathology. Humans are less likely to make such superficial mistakes, since they use features that are grounded on medical science. It is therefore important to know whether DNNs use different features than humans. Towards this end, we propose a framework for comparing human and machine perception in medical diagnosis. We frame the comparison in terms of perturbation robustness, and mitigate Simpson's paradox by performing a subgroup analysis. The framework is demonstrated with a case study in breast cancer screening, where we separately analyze microcalcifications and soft tissue lesions. While it is inconclusive whether humans and DNNs use different features to detect microcalcifications, we find that for soft tissue lesions, DNNs rely on high frequency components ignored by radiologists. Moreover, these features are located outside of the region of the images found most suspicious by radiologists. This difference between humans and machines was only visible through subgroup analysis, which highlights the importance of incorporating medical domain knowledge into the comparison.

Reducing False-Positive Biopsies using Deep Neural Networks that Utilize both Local and Global Image Context of Screening Mammograms.

Breast cancer is the most common cancer in women, and hundreds of thousands of unnecessary biopsies are done around the world at a tremendous cost. It is crucial to reduce the rate of biopsies that turn out to be benign tissue. In this study, we build deep neural networks (DNNs) to classify biopsied lesions as being either malignant or benign, with the goal of using these networks as second readers serving radiologists to further reduce the number of false-positive findings. We enhance the performance of DNNs that are trained to learn from small image patches by integrating global context provided in the form of saliency maps learned from the entire image into their reasoning, similar to how radiologists consider global context when evaluating areas of interest. Our experiments are conducted on a dataset of 229,426 screening mammography examinations from 141,473 patients. We achieve an AUC of 0.8 on a test set consisting of 464 benign and 136 malignant lesions.

An artificial intelligence system for predicting the deterioration of COVID-19 patients in the emergency department.

During the coronavirus disease 2019 (COVID-19) pandemic, rapid and accurate triage of patients at the emergency department is critical to inform decision-making. We propose a data-driven approach for automatic prediction of deterioration risk using a deep neural network that learns from chest X-ray images and a gradient boosting model that learns from routine clinical variables. Our AI prognosis system, trained using data from 3661 patients, achieves an area under the receiver operating characteristic curve (AUC) of 0.786 (95% CI: 0.745-0.830) when predicting deterioration within 96 hours. The deep neural network extracts informative areas of chest X-ray images to assist clinicians in interpreting the predictions and performs comparably to two radiologists in a reader study. In order to verify performance in a real clinical setting, we silently deployed a preliminary version of the deep neural network at New York University Langone Health during the first wave of the pandemic, which produced accurate predictions in real-time. In summary, our findings demonstrate the potential of the proposed system for assisting front-line physicians in the triage of COVID-19 patients.

An artificial intelligence system for predicting the deterioration of COVID-19 patients in the emergency department.

During the coronavirus disease 2019 (COVID-19) pandemic, rapid and accurate triage of patients at the emergency department is critical to inform decision-making. We propose a data-driven approach for automatic prediction of deterioration risk using a deep neural network that learns from chest X-ray images and a gradient boosting model that learns from routine clinical variables. Our AI prognosis system, trained using data from 3,661 patients, achieves an area under the receiver operating characteristic curve (AUC) of 0.786 (95% CI: 0.745-0.830) when predicting deterioration within 96 hours. The deep neural network extracts informative areas of chest X-ray images to assist clinicians in interpreting the predictions and performs comparably to two radiologists in a reader study. In order to verify performance in a real clinical setting, we silently deployed a preliminary version of the deep neural network at New York University Langone Health during the first wave of the pandemic, which produced accurate predictions in real-time. In summary, our findings demonstrate the potential of the proposed system for assisting front-line physicians in the triage of COVID-19 patients.

Infection mechanism of biofilm-forming Staphylococcus aureus on indwelling foreign materials in mice.

Indwelling foreign-body infections are a critical medical problem, especially in immunocompromised patients. To examine the pathogenicity of biofilm-forming bacteria settling on foreign materials, mice implanted with plastic discs were infected with Staphylococcus aureus. After opening a wide subcutaneous pocket on the dorsal side of mice with or without temporal leukocytopenia, a plastic sheet was placed in the left subcutaneous space; subsequently, bacteria in a planktonic state were dispersed over the subcutaneous space. Bacterial numbers were examined 7 days after inoculation. In subcutaneous tissue on the right, S. aureus was found only in leukocytopenic mice. Meanwhile, bacteria were detected on the plastic and neighbouring tissue in both leukocytopenic and normal mice; however, colony-forming analysis indicated that leukocytopenic mice possessed significantly more bacteria. Tissue reaction against bacteria was pathologically examined. Invading S. aureus induced severe inflammation. In transient leukocytopenic mice, bacterial microcolonies formed on the plastic as well as in the developed necrotic tissue - both of which were shielded from inflammatory cell infiltration - result in bacteraemia. These results indicate that biofilm-forming S. aureus settling on indwelling foreign material are tolerant against host immunity and assault neighbouring tissue, which may lead to chronic wound infection.

A new sternum elevator reduces severe complications during minimally invasive repair of the pectus excavatum.

INTRODUCTION: Although the Nuss procedure for pectus excavatum is widely employed, a variety of complications have been reported with relatively high frequency; those that involve cardiac and pericardial injuries can be life threatening. To reduce such dangers, we present here a newly developed sternal elevator. MATERIALS AND METHODS: The elevator is horseshoe shaped. Its elevator side has the same curvature as a Nuss introducer, so that interference between devices is minimal and no extra skin incision is needed for the elevator insertion. The elevator holds the sternum forward and enlarges the retrosternal space for safer passage of thoracoscopically guided introducer. RESULTS: The authors have used the elevator for 61 pectus excavatum cases between March 2004 and December 2009 without any major complications. The entire process of substernal tunneling was endoscopically observed, which eliminated any blunt and blind dissection, even in a significantly depressed funnel chest case. With the device, the sternum was effectively elevated again for the placement of the second plate in 30 cases. CONCLUSION: Our newly developed sternum elevator makes the Nuss procedure safer and more affordable without introducing any extra scarring.

Bilayer reconstruction for parry-romberg syndrome: using a free circumflex scapular artery-based adipofascial flap for both the buccal fat pad and subcutaneous fat.

PURPOSE: When using a free flap to reconstruct a facial deformity caused by Romberg's disease, it is important to prevent the flap from sagging after the operation. We report a new method of reconstructive surgery using a free subscapular adipofascial flap to prevent this problem. PATIENTS AND METHODS: Three female patients (ages 27, 28, and 34 years) with Parry-Romberg syndrome underwent microsurgical free scapular flap transfer for buccal defects. This operation requires making a gingivobuccal sulcus incision and forming a pocket for buccal fat reconstruction by dissecting over the periosteum of the maxillary bone. Preauricular and submandibular incisions are made to create a subcutaneous pocket for flap transfer. After the subscapular flap is elevated, we use its angiogram to observe its vascular pattern. The flap is separated to preserve the main blood vessels horizontal lower branches. The subcutaneous adipose tissue layer uses the horizontal branch, and the buccal fat pad layer the lower branch. METHODS: After the operations, the adipofascial flaps were in good condition and without postoperative complications. A half year after the first operation, revisional surgery was performed for one patient. No cases showed no sagging of the cheek, and in every case the overall appearance of the buccal region improved significantly. DISCUSSION AND CONCLUSION: Reconstruction of the buccal fat pad and subcutaneous adipose tissue using a free bilobed adipofascial flap nourished by the circumflex scapular artery, returned the adiposal tissue to its normal position, assuring more natural facial contours.

Fluorescence emission and polarization for analyzing binding of ruthenium metalloglycocluster to lectin and tetanus toxin C-fragment.

We have designed and synthesized ruthenium complexes bearing clustered galactose Ru(bpy-2Gal)(3) and glucose Ru(bpy-2Glc)(3). Changes in fluorescence emission (FE) and fluorescence polarization (FP) of the metalloglycoclusters were measured by adding each lectin (peanut agglutinin (PNA), Ricinus communis agglutinin 120 (RCA), concanavalin A (ConA), or wheat germ agglutinin (WGA)) or tetanus toxin c-fragment (TCF). Following the addition of PNA and ConA, the FE spectra of Ru(bpy-2Gal)(3) and Ru(bpy-2Glc)(3) showed new emission peaks, respectively. In addition, Ru(bpy-2Gal)(3) and Ru(bpy-2Glc)(3) exclusively increased the FP values by addition of PNA and ConA. Since other combinations of the metalloglycoclusters and lectin caused little change, specific bindings of galactose to PNA and glucose to ConA were confirmed by the FE and FP measurement. From the FP analyses, the dissociation constants (K(d)) of Ru(bpy-2Gal)(3) to PNA and Ru(bpy-2Glc)(3) to ConA were calculated to be ca. 6.1 x 10(-6) M and 1.8 x 10(-5) M. Furthermore, the FP analyses proved specific binding of Ru(bpy-2Gal)(3) to TCF.

Cell-cycle dependent dynamic change of 26S proteasome distribution in tobacco BY-2 cells.

The 26S proteasome is known to play pivotal roles in cell-cycle progression in various eukaryotic cells; however, little is known about its role in higher plants. Here we report that the subcellular distribution of the 26S proteasome is dynamically changed in a cell-cycle dependent manner in tobacco BY-2 cells as determined by immunostaining with anti-Rpn10 (a regulatory PA700 subunit) and anti-20S catalytic proteasome antibodies. The 26S proteasome was found to localize not only in nuclear envelopes and mitotic spindles but also in preprophase bands (PPBs) and phragmoplasts appearing in G(2) and M phases, respectively. MG132, a proteasome inhibitor, exclusively caused cell-cycle arrest not only at the metaphase but also the early stage of PPB formation at the G(2) phase and the collapse of the phragmoplast, which seems to be closely related to proteasome distribution in the cells.