Engraftment of allogeneic hematopoietic progenitor cells with purine analog-containing chemotherapy: harnessing graft-versus-leukemia without myeloablative therapy

The immune-mediated graft-versus-leukemia effect is important to prevent relapse after allogeneic progenitor cell transplantation. This process requires engraftment of donor immuno-competent cells. The objective of this study was to assess the feasibility of achieving engraftment of allogeneic peripheral blood or bone marrow progenitor cell after purine analog containing nonmyeloablative chemotherapy. Patients with advanced leukemia or myelodysplastic syndromes (MDS) who were not candidates for a conventional myeloablative therapy because of older age or organ dysfunction were eligible. All patients had an HLA-identical or one-antigen-mismatched related donor. Fifteen patients were treated (13 with acute myeloid leukemia and 2 with MDS). The median age was 59 years (range, 27 to 71 years). Twelve patients were either refractory to therapy or beyond first relapse. Eight patients received fludarabine at 30 mg/m2/d for 4 days with idarubicin at 12 mg/m2/d for 3 days and ara-c at 2 g/m2/d for 4 days (n = 7) or melphalan at 140 mg/m2/d (n = 1). Seven patients received 2-chloro-deoxyadenosine at 12 mg/m2/d for 5 days and ara-C 1 at g/m2/d for 5 days. Thirteen patients received allogeneic peripheral blood stem cells and 1 received bone marrow after chemotherapy. Graft-versus-host disease (GVHD) prophylaxis consisted of cyclosporine and methyl-prednisolone. Treatment was generally well tolerated, with only 1 death from multiorgan failure before receiving stem cells. Thirteen patients achieved a neutrophil count of greater than 0.5 x 10(9)/L a median of 10 days postinfusion (range, 8 to 17 days). Ten patients achieved platelet counts of 20 x 10(9)/L a median of 13 days after progenitor cell infusion (range, 7 to 78 days). Eight patients achieved complete remissions (bone marrow blasts were < 5% with neutrophil recovery and platelet transfusion independence) that lasted a median of 60 days posttransplantation (range, 34 to 170+ days). Acute GVHD grade > or = 2 occurred in 3 patients. Chimerism analysis of bone marrow cells in 6 of 8 patients achieving remission showed > or = 90% donor cells between 14 and 30 days postinfusion, and 3 of 4 patients remaining in remission between 60 and 90 days continued to have > or = 80% donor cells. We conclude that purine analog-containing nonmyeloablative regimens allow engraftment of HLA-compatible hematopoietic progenitor cells. This approach permits us to explore the graft-versus-leukemia effect without the toxicity of myeloablative therapy and warrants further study in patients with leukemia who are ineligible for conventional transplantation with myeloablative regimens either because of age or concurrent medical conditions.     

Structuring communication software for quality-of-serviceguarantees

We propose architectural mechanisms for structuring host communication software to provide QoS guarantees. We present and evaluate a QoS sensitive communication subsystem architecture for end hosts that provides real time communication support for generic network hardware. This architecture provides services for managing communication resources for guaranteed QoS (real time) connections, such as admission control, traffic enforcement, buffer management, and CPU and link scheduling. The architecture design is based on three key goals: maintenance of QoS guarantees on a per connection basis, overload protection between established connections, and fairness in delivered performance to best effort traffic. Using this architecture we implement real time channels, a paradigm for real time communication services in packet switched networks. The proposed architecture features a process per channel model that associates a channel handler with each established channel. The model employed for handler execution is one of “cooperative” preemption, where an executing handler yields the CPU to a waiting higher priority handler at well defined preemption points. The architecture provides several configurable policies for protocol processing and overload protection. We present extensions to the admission control procedure for real time channels to account for cooperative preemption and overlap between protocol processing and link transmission at a sending host. We evaluate the implementation to demonstrate the efficacy with which the architecture maintains QoS guarantees on outgoing traffic while adhering to the stated design goals     

Stability and throughput analysis of unslotted CDMA‐ALOHA with finite number of users and code sharing

We consider a system comprising a finite number of nodes, with infinite packet buffers, that use unslotted ALOHA with Code Division Multiple Access (CDMA) to share a channel for transmitting packetised data. We propose a simple model for packet transmission and retransmission at each node, and show that saturation throughput in this model yields a sufficient condition for the stability of the packet buffers; we interpret this as the capacity of the access method. We calculate and compare the capacities of CDMA‐ALOHA (with and without code sharing) and TDMA‐ALOHA; we also consider carrier sensing and collision detection versions of these protocols. In each case, saturation throughput can be obtained via analysis of a continuous time Markov chain. Our results show how saturation throughput degrades with code‐sharing. Finally, we also present some simulation results for mean packet delay. Our work is motivated by optical CDMA in which “chips” can be optically generated, and hence the achievable chip rate can exceed the achievable TDMA bit rate which is limited by electronics. Code sharing may be useful in the optical CDMA context as it reduces the number of optical correlators at the receivers. Our throughput results help to quantify by how much the CDMA chip rate should exceed the TDMA bit rate so that CDMA‐ALOHA yields better capacity than TDMA‐ALOHA. This revised version was published online in June 2006 with corrections to the Cover Date.     

Designing content distribution networks for optimal cost and performance

One strategy for alleviating excess latency (delay) in the Internet is the caching of web content at multiple locations. This reduces the number of hops necessary to reach the desired content. This strategy is used for web content such as html pages, images, streaming video, and Internet radio. The network of servers which store this content, and the collections of objects stored on each server, is called a content distribution network (CDN). In order to optimally design a CDN, given a network topology with available server storage capacity at various points in the network, one must decide which object collections to place on each server in order to achieve performance or cost objectives. The placements must be within the storage limits of the servers and must reflect the request patterns for each collection of objects to be cached. Researchers have suggested formulations for the CDN problem which address performance by minimizing latency (the average number of hops is a commonly accepted measure of latency) from client to content, or formulations that focus on minimizing cost of storage and/or bandwidth. In this research, we develop a model which allows for the simultaneous treatment of performance and cost, present examples to illustrate the application of the model and perform a detailed designed experiment to gain insights into cost/hops tradeoff for a variety of network parameters.     

A role for isothiocyanates in plant resistance against the specialist herbivore Pieris rapae

We experimentally reanalyzed the classic interaction between Pieris rapae, a specialist lepidopteran herbivore, and isothiocyanates (mustard oils) that are characteristic phytochemicals of the Brassicaceae. Previous investigations have suggested that P. rapae is unaffected by isothiocyanates. Using whole plants, root extracts, and a microencapsulated formulation of allyl isothiocyanate, we now show that isothiocyanates reduce herbivore survival and growth, and increase development time, each in a dose-dependent manner. Neither the substrate allyl glucosinolate, nor myrosinase, the enzyme that results in the breakdown of glucosinolates, negatively affected P. rapae. Thus, we present strong evidence for a role for isothiocyanates in plant resistance against the specialist herbivore P. rapae.     

Crystallization kinetics of bulk amorphous Se80−xSbxTe20

Crystallization kinetics of the Se_80–x Sb _x Te₂₀ (0x9) alloys have been studied using differential scanning calorimetry. The activation energies for the glass transition and that for crystallization have been determined from the heating rate dependence of the glass transition temperature and the peak crystallization temperature. The results have been analysed using the modified Kissinger's and Matusita's equations for the non-isothermal crystallization of materials. The variation of glass transition temperature with composition suggests that a small amount of Sb ( 4 at %) leads to an increase in the chain length of Se-Te, whereas further increase in Sb atomic per cent increases the number of Se-Te chains in the alloys.     

Modeling finite buffer effects on TCP traffic over an IEEE 802.11 infrastructure WLAN

The network scenario is that of an infrastructure IEEE 802.11 WLAN with a single AP with which several stations (STAs) are associated. The AP has a finite size buffer for storing packets. In this scenario, we consider TCP-controlled upload and download file transfers between the STAs and a server on the wireline LAN (e.g., 100 Mbps Ethernet) to which the AP is connected. In such a situation, it is well known that because of packet losses due to finite buffers at the AP, upload file transfers obtain larger throughputs than download transfers. We provide an analytical model for estimating the upload and download throughputs as a function of the buffer size at the AP. We provide models for the undelayed and delayed ACK cases for a TCP that performs loss recovery only by timeout, and also for TCP Reno. The models are validated in comparison with NS2 simulations.     

Automatic recognition of handwritten medical forms for search engines

A new paradigm, which models the relationships between handwriting and topic categories, in the context of medical forms, is presented. The ultimate goals are: (1) a robust method which categorizes medical forms into specified categories, and (2) the use of such information for practical applications such as an improved recognition of medical handwriting or retrieval of medical forms as in a search engine. Medical forms have diverse, complex and large lexicons consisting of English, Medical and Pharmacology corpus. Our technique shows that a few recognized characters, returned by handwriting recognition, can be used to construct a linguistic model capable of representing a medical topic category. This allows (1) a reduced lexicon to be constructed, thereby improving handwriting recognition performance, and (2) PCR (Pre-Hospital Care Report) forms to be tagged with a topic category and subsequently searched by information retrieval systems. We present an improvement of over 7% in raw recognition rate and a mean average precision of 0.28 over a set of 1,175 queries on a data set of unconstrained handwritten medical forms filled in emergency environments. This work was supported by the National Science Foundation.     

Unconstrained handwritten document retrieval

With the ever-increasing growth of the World Wide Web, there is an urgent need for an efficient information retrieval system that can search and retrieve handwritten documents when presented with user queries. However, unconstrained handwriting recognition remains a challenging task with inadequate performance thus proving to be a major hurdle in providing robust search experience in handwritten documents. In this paper, we describe our recent research with focus on information retrieval from noisy text derived from imperfect handwriting recognizers. First, we describe a novel term frequency estimation technique incorporating the word segmentation information inside the retrieval framework to improve the overall system performance. Second, we outline a taxonomy of different techniques used for addressing the noisy text retrieval task. The first method uses a novel bootstrapping mechanism to refine the OCR’ed text and uses the cleaned text for retrieval. The second method uses the uncorrected or raw OCR’ed text but modifies the standard vector space model for handling noisy text issues. The third method employs robust image features to index the documents instead of using noisy OCR’ed text. We describe these techniques in detail and also discuss their performance measures using standard IR evaluation metrics.