Thesis (Ph. D.)--Harvard--Massachusetts Institute of Technology Division of Health Sciences and Technology, 1998. / Includes bibliographical references (p. 157-170). / by Thomas James Mullen. / Ph.D.

Thesis (Ph. D.)--Harvard-MIT Division of Health Sciences and Technology, 2006. / Includes bibliographical references (p. 129-140). / Investigating the neural mechanisms underlying the perception of the pitch of harmonic complex tones is of great importance for many reasons. Changes in pitch convey melody in music, and the superposition of different pitches is the basis for harmony. Pitch has an important role in speech, where it carries prosodic features and information about speaker identity. Pitch plays a major role in auditory scene analysis: differences in pitch are a major cue for sound source segregation, while frequency components that share a common fundamental frequency (FO) tend to be grouped into a single auditory object. In psychophysics, a positive correlation is commonly observed between the estimated "resolvability" of individual harmonics of complex tones, assumed to depend primarily on the frequency selectivity of the cochlea, and the strength of the corresponding pitch percepts. In this study, possible neural codes for the pitch of harmonic complex tones were investigated in the auditory nerve of anesthetized cats, with particular focus on their dependence on cochlear frequency selectivity, which was measured directly using both complex tones and band-reject noise. A "rate-place" representation of pitch, based on cues to peripherally-resolved harmonics in profiles of average discharge rate along tonotopically-arranged neurons, was compared to a "temporal" representation, based on periodicity cues in the distributions of interspike intervals of the entire auditory nerve. / (cont.) Although both representations were viable in the range of FOs of cat vocalizations, neither was entirely satisfactory in accounting for psychophysical data. The rate-place representation degraded rapidly with increasing stimulus level and could not account for the upper limit of the perception of the pitch of missing-F0 in humans, while the interspike-interval representation could not predict the correlation between psychophysical pitch salience and peripheral harmonic resolvability. Therefore, we tested an alternative, "spatio-temporal" representation of pitch, where cues to the resolved harmonics arise from the spatial pattern in the phase of the basilar membrane motion. The spatio-temporal representation was relatively stable with level and was consistent with an upper limit for the pitch of missing-F0, thus becoming the strongest candidate to explain several major human pitch perception phenomena. / by Leonardo Cedolin. / Ph.D.

Thesis (Sc. D.)--Harvard-MIT Division of Health Sciences and Technology, 2004. / Includes bibliographical references (p. 307-314). / Clinical gait analysis currently involves either an expensive analysis in a motion laboratory, using highly accurate, if cumbersome, kinematic systems, or a qualitative analysis with a physician or physical therapist making visual observations. There is a need for a low cost device that falls in between these two methods, and can provide quantitative and repeatable results. In addition, continuous monitoring of gait would be useful for real-time physical rehabilitation. To free patients from the confines of a motion laboratory, this thesis has resulted in a wireless wearable system capable of measuring many parameters relevant to gait analysis. The extensive sensor suite includes three orthogonal accelerometers, and three orthogonal gyroscopes, four force sensors, two bi-directional bend sensors, two dynamic pressure sensors, as well as electric field height sensors. The "GaitShoe" was built to be worn on any shoes, without interfering with gait, and was designed to collect data unobtrusively, in any environment, and over long periods of time. Subject testing of the GaitShoe was carried out on ten healthy subjects with normal gait and five subjects with Parkinson's disease. The calibrated sensor outputs were analyzed, and compared to results obtained simultaneously from The Massachusetts General Hospital Biomotion Lab; the GaitShoe proved highly capable of detecting heel strike and toe off, as well as estimating orientation and position of the subject. A wide variety of features were developed from the calibrated sensor outputs, for use with standard pattern recognition techniques to classify the gait of the subject. The results of the classification demonstrated the ability of the GaitShoe to identify the subjects with / (cont.) Parkinson's disease, as well as individual subjects. Real-time feedback methods were developed to investigate the feasibility of using the continuous monitoring of gait for physical therapy and rehabilitation. / by Stacy J. Morris. / Sc.D.

Thesis (S.M.)--Harvard-MIT Division of Health Sciences and Technology, 2007. / Includes bibliographical references. / Vaso-occlusive crises cause most of the morbidity and mortality associated with sickle cell disease. The proximal causes of these occlusive events are not well understood. The risks and consequences of vaso-occlusion however are clear. Ten percent of sickle cell disease patients will have a stroke by the age of 20. Two thirds of sickle cell disease patients require more than one hospitalization per year for treatment of pain crises. The flow behavior of blood samples from sickle cell patients was studied in an artificial microfluidic environment. This microfluidic environment allowed modulation of the hydrostatic pressure causing flow, the ambient oxygen concentration, and the vascular channel geometry. A range of blood samples was evaluated by selecting specimens with various hematocrits and concentrations of sickle hemoglobin. Velocity profiles were calculated following sudden changes in oxygen concentration. From these profiles, it was possible to create a phase space of vaso-occlusion in the artificial microfluidic environment. This phase space characterizes the environmental conditions in which sickle cell blood will stop flowing within a given interval of time. / (cont.) This work is a first step in characterizing the inter-relationships between some of the control parameters governing vaso-occlusion: pressure, oxygen concentration, channel size, hematocrit, and sickle hemoglobin concentration. This artificial device enables a quantification of the effect of a clinical therapy, red blood cell exchange, as performed on an actual sickle cell patient. Additionally, three sample small molecules known to alter rates of sickle hemoglobin polymerization were evaluated for their ability to perturb the tendency of sickle cell blood to stop flowing. These results suggest a possible application of this technique to the diagnosis and monitoring of sickle cell patients as well as to the investigation of new regimens of existing treatments and altogether novel therapies. / by John M. Higgins. / S.M.

Thesis (S.M.)--Harvard-MIT Division of Health Sciences and Technology, 2009. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 68-69). / Corporate spin-outs have become more frequent in the contemporary business environment as an alternate source of risk diversification and value creation for both the parent and external investors. Once established, corporate spin-outs are often perceived to be of higher quality than their counterparts in the industry; previous studies have shown that they tend to receive higher valuations in financing, faster financing and higher preference by prestigious Wall Street investment banks when they decide to go public. The primary objective of this thesis was to compare the net proceeds associated with successful liquidity events (IPO or M&A) for US based therapeutic-focused corporate spin-outs to industry averages and test the hypothesis that corporate spin-outs generate superior returns. A database containing information on 186 corporate spin-outs within the life sciences (founded from 1990 - present) was generated for the purpose of testing this hypothesis. Net proceeds from corporate spin-out liquidity events were compared to median net proceeds of all biotech/pharmaceutical liquidity events for a given vintage year and type of liquidity event (IPO vs. M&A). Liquidity events were observed with a / (cont.) higher frequency than overall industry averages. Results indicated that net IPO proceeds were similar to industry averages, while M&A proceeds were above the median vintage year value for every case observed. When normalizing by the most advanced clinical stage program, a similar trend was observed in three of the five cases. In addition, internal rate of return (IRR) and cash on cash exit multiple for Series A investors was substantially higher in corporate spin-outs than industry averages. In order to understand why acquisitions of corporate spin-outs appeared to generate sizable excess returns relative to industry averages, qualitative interviews were conducted with former executives involved in these transactions. Key insights from these interviews indicate that a seasoned management team, prestige of parent company, high quality syndicate of investors, clinically proven technology and a clear regulatory path to approval are all elements that help drive excess valuations of corporate spin-outs in the life sciences. We conclude that corporate spin-outs do generate superior returns through M&A exits compared to venture-backed start ups, while proceeds from IPO's were similar to case controls. / by Michael Robert Magnani. / S.M.

Thesis (M. Eng.)--Harvard-MIT Division of Health Sciences and Technology, 2008. / Includes bibliographical references (leaves 51-56). / Cell-based high-throughput screening is emerging as a disruptive technology in drug discovery; however, massively parallel electrical assaying of neurons and cardiomyocites has until now been prohibitively expensive. To address this limitation, we developed a scalable, all-organic 3D microelectrode array technology. The cheap, disposable arrays would be integrated into a fixed stimulation and imaging setup, potentially amenable to automated handling and data analysis. A combination of activity-dependent plasticity, made possible by independent control of up to 64 stimulating electrodes, and, eventually, of substrate chemical patterning would be employed to constrain the neuronal culture network connectivity. In order to ensure longterm survival of the cultures, a bottom feeder layer of glial cells would be grown. In addition to high-throughput screening application, the polymeric microelectrode arrays and integrated stimulation systems were designed to allow the long-term study of synaptic plasticity, combining excellent long-term culture capabilities with a unique ability to independently control each electrode stimulation pattern. The resulting activity could be monitored optically, e,g, with calcium or voltage sensitive dyes, and the images could be stored and processed (possibly even in real time) within the same environment (LabView) as the stimulator. To fabricate the polymeric microelectrode array, we prepare a multilayered mask substrate, by reversibly bonding together two sheets of implant-grade polydimethylsiloxane (PDMS) sheets, with or without a glass coverslip between them. Thanks to PDMS self-adhesive properties the various layers are held together stably but reversibly. The mask is then laser-patterned, using either a standard CO2 laser or a 193 nm excimer laser. / (cont.) The mask can then be adhered onto a glassy carbon or ITO electrode, and polypyrrole, doped with either hyaluronic acid or sodium dodecylbenzesulfonic acid, can be electrodeposited through it. Finally, the construct is removed from the deposition bath and the upper, sacrificial mask layer carefully peeled away. This fabrication method allows exquisite control overall 3D electrode geometry, is suitable to produce structures between one and several hundred micrometers in diameter, either filled or tubular, and scales extremely well, so that, for example, 384 by 64 electrodes arrays can be patterned in just a few minutes and grown in the same time as a single array. / by Giovanni Talei Franzesi. / M.Eng.

Thesis (Ph. D.)--Harvard-MIT Division of Health Sciences and Technology, 2011. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 243-260). / The primary function of the heart is to pump blood at a sufficient rate to ensure perfusion of all the organs. This vital task is achieved in large part by controlling the rate of cardiac contractions, which are initiated by cells in the sinoatrial node, the "pacemaker" of the heart. The oscillation rate of these spontaneously active cells is tightly regulated by the sympathetic and parasympathetic branches of the autonomic nervous system. Our understanding of sinoatrial node cell function has been greatly advanced by experimental and modeling efforts that quantitatively describe the numerous ionic currents responsible for the cell's spontaneous depolarization and generation of the action potential. Several models have also explored the effect of sympathetic and parasympathetic activity on specific ion channels and have reproduced the classic slowing and acceleration phenomena. However, a complete model of this interaction does not exist: current models lack the ability to simulate simultaneous sympathetic and parasympathetic activation or to reproduce heart rate dynamics in response to time-varying autonomic inputs. We addressed this need by constructing a bottom-up model of sinoatrial node cell regulation by the autonomic nervous system, with a focus on reproducing the full range of heart rates observed under simultaneous sympathetic and parasympathetic nerve stimulation, as well as the dynamic heart rate response to steps in sympathetic or parasympathetic stimulation rate. In constructing our model, we consolidate a large body of experimental data in a consistent mathematical framework. The model comprises 57 nonlinear coupled ordinary differential equations based on first principles and the current mechanistic understanding of the component reactions, fits well all the experimental data used to build the model, and reproduces high-level features of the system that were not explicitly fit when building the model. The detailed nature of the model also allows numerous conclusions to be drawn about the mechanisms of heart rate control. A better understanding of these mechanisms in health and disease may enable the development of better diagnostics for cardiovascular disease and more targeted drug design. We also identified a number of limitations in the present model that can be refined through further experimental and numerical efforts. / by Danilo Šćepanović. / Ph.D.

Thesis (Ph. D.)--Harvard-MIT Program in Health Sciences and Technology, 2012. / Cataloged from PDF version of thesis. / Includes bibliographical references. / Myotonic dystrophy (DM) is the most common form of adult onset muscular dystrophy, affecting more than 1 in 8000 individuals globally. The symptoms of DM are multi-systemic and include myotonia, severe muscle wasting, cardiac arrhythmias, cataracts, gastrointestinal dysfunction, and cognitive deficits. DM is caused by the expansion of CTG or CCTG repeat sequences expressed in noncoding portions of RNA, which sequester or activate RNA splicing factor proteins, leading to widespread deleterious changes in transcriptome isoform usage. We developed a method for studying transcriptomes, RNAseq, which provides a high resolution, digital inventory of gene and isoform expression. By applying RNAseq to human tissues and cell lines, we discovered that essentially 92-94% of all human genes are alternatively spliced, 86% of them with a minor isoform frequency 15% or more. We found that the majority of alternative splicing and alternative polyadenylation and cleavage events are tissue-regulated, and that patterns of these RNA processing events are strongly correlated across tissues, implicating protein factors that may regulate both types of events. We applied this method towards the goal of identifying transcriptome changes occurring in DM, focusing on the Muscleblind-like (MBNL) family of RNA binding proteins, which are functionally inactivated by CUG or CCUG repeats. Using RNAseq to profile tissues and cells depleted of MBNLs, we found that MBNL1 and MBNL2 co-regulate hundreds of redundant targets. MBNL1 UV cross-linking and immunoprecipitation, followed by sequencing (CLIPseq), was used to identify the in vivo transcriptome-wide binding locations of MBNL1, and facilitated the construction of a context-dependent RNA map for MBNL1 splicing regulation. Extensive 3' UTR binding of MBNL1 was found to localize mRNAs to membrane compartments of mouse myoblasts, suggesting a new global function for MBNLs, and additional mechanisms by which MBNL depletion can lead to DM symptoms. / by Eric T. Wang. / Ph.D.

Thesis: Ph. D. in Biomedical Engineering, Harvard-MIT Program in Health Sciences and Technology, 2017. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 213-227). / Progressive bulbar motor deterioration resulting from amyotrophic lateral sclerosis (ALS) leads to speech impairment. Despite the devastating consequences of speech impairment to life quality, few options are available to objectively assess speech motor involvement. The overarching goal of this research was to derive objective measures of speech acoustics that can be used to support clinical decision making. To achieve this goal, we obtained 121 speech samples from 33 patients with ALS who repeated the phrase "Buy Bobby a puppy" five times in succession. In total, 342 acoustic features were semi-automatically extracted from each speech recording. Pearson correlations were computed between each feature and three metrics of overall speech severity: sentence intelligibility, speaking rate, and communication efficiency. The findings were grounded within a physiologic framework where acoustic features were grouped into one of three domains that when combined, were hypothesized to broadly characterize articulatory performance: articulatory specification, articulatory coupling, and articulatory consistency. To obtain the most accurate prediction of ALS with the features we extracted, we compared two machine learning algorithms: linear regression and random forest. In shuffle-split cross-validation, the strongest mean Pearson correlations we obtained between actual and predicted intelligibility, speaking rate, and communication efficiency were 0.67, 0.74, and 0.77, respectively (SD=0.077, 0.050, and 0.059, respectively). Of the three domains, the specificity features were the most strongly associated with intelligibility impairments (mean r=0.68), and coupling was the most strongly associated with slower speaking rate (mean r=0.73). Specificity and coupling yielded similar performances in communication efficiency prediction. Other contributions of this thesis are that it is the first to implement a framework of dysarthric speech in terms of three domains: specification, coupling, and consistency; the first to validate automated formant tracking in dysarthric speech; and the first to perform an in-depth investigation into physiologically-inspired acoustic features that describe articulatory impairments of patients with ALS. Novel findings include the presence of abnormal formant coupling patterns, which may suggest greater tonguejaw coupling, in patients with more severe dysarthria due to ALS. Areas of future research involve further feature discovery, improved analysis methods, and a deeper understanding of relations to articulatory kinematics. / by Rachelle L. Horwitz-Martin. / Ph. D. in Biomedical Engineering

Thesis (S.M.)--Harvard-MIT Division of Health Sciences and Technology, 2005. / Includes bibliographical references (leaves 55-57). / Regulation of gene expression occurs largely through the binding of sequence- specific transcription factors (TFs) to genomic DNA binding sites (BSs). This thesis presents a rigorous scoring scheme, implemented as a C program termed "ModuleFinder", that evaluates the likelihood that a given genomic region is a cis regulatory module (CRM) for an input set of TFs according to its degree of: (1) homotypic site clustering; (2) heterotypic site clustering; and (3) evolutionary conservation across multiple genomes. Importantly, ModuleFinder obtains all parameters needed to appropriately weight the relative contributions of these sequence features directly from the input sequences and TFBS motifs, and does not need to first be trained. Using two previously described collections of experimentally verified CRMs in mammals as validation datasets, we show that ModuleFinder is able to identify CRMs with great sensitivity and specificity. We also evaluated ModuleFinder on a set of DNA binding site data for the human TFs Hepatocyte Nuclear Factor HNF1 [alpha], HNF4 [alpha] and HNF6 and compared its performance with logistic regression and neural network models. / by Fangxue He. / S.M.