Vishnu Reddy

Cardiac involvement by metastatic neoplasms is relatively uncommon and usually occurs with widely disseminated disease. Ninety-five cases with cardiac metastases from autopsies performed over a 14-year period (1974-1987) at Loyola University Medical Center are reviewed. During this period, 3314 autopsies were performed with an average annual autopsy rate of 35%. In 806 (24.3%), a malignant disease was found, and in 95 (11.8%), there was cardiac involvement by tumor. The most common malignancies encountered in order of decreasing frequency were lung, lymphoma, breast, leukemia, stomach, melanoma, liver, and colon. Although the percentage of cardiac metastasis compares favorably with previous reports in the literature, an identical rate was present during both halves of the 14-year period studied. Improved diagnostic capabilities and treatment protocols in recent years have apparently not significantly affected the incidence, distribution, or patterns of metastatic spread to the heart.

This paper studied automatic identification of malaria infected cells using deep learning methods. We used whole slide images of thin blood stains to compile an dataset of malaria-infected red blood cells and non-infected cells, as labeled by a group of four pathologists. We evaluated three types of well-known convolutional neural networks, including the LeNet, AlexNet and GoogLeNet. Simulation results showed that all these deep convolution neural networks achieved classification accuracies of over 95%, higher than the accuracy of about 92% attainable by using the support vector machine method. Moreover, the deep learning methods have the advantage of being able to automatically learn the features from the input data, thereby requiring minimal inputs from human experts for automated malaria diagnosis.

We describe a woman with severe neutropenia and dependency on red blood cell transfusions who had previously undergone Billroth II surgery and whose bone marrow (BM) showed morphologic characteristics typical of myelodysplastic syndrome (MDS) with ringed sideroblasts. She had transient reversal of anemia and severe neutropenia after therapy with erythropoietin and granulocyte colony-stimulating factor. Because of relapse while receiving growth factors, the patient was referred for allogeneic BM transplantation. A pretransplantation nutritional evaluation revealed severe copper deficiency, and her hematologic abnormalities resolved fully with copper therapy. This case shows that copper deficiency should be an integral part of the differential diagnosis of sideroblastic MDS, even in patients not requiring parenteral nutrition.

In this paper, we first study the signal cancellation and interference rejection effects of the optimum (constrained least squares or minimum variance) beamformer in the presence of partially and fully correlated interfering sources. In particular, we derive expressions for the output power and the gain in the interference direction of the beamformer in terms of the source powers, correlation, and the sensor noise power, and show quantitatively the penalties arising from increasing correlation in several scenarios of interest. Next, we show that spatial smoothing progressively decorrelates the sources at a rate that depends on the spacing and directions of the sources, and thus relate the degree of smoothing to the improvement in signal cancellation and interference rejection behavior provided by spatial smoothing. Results of computer simulations are included to support our analysis.

We summarize here a case involving a fatal sickle cell crisis in a patient with previously very mild hemoglobin sickle cell (Hb SC) disease. With granulocyte colony-stimulating factor (G-CSF) priming for peripheral stem cell collection, a crisis developed, followed by death within 36 hours. The case