Annotated Curriculum Vitae - Lawrence M. Lifshitz
Education
Professional History
1987-2002: Assistant Professor of Physiology and Nuclear Medicine and member of the Biomedical Imaging Group, University of Massachusetts Medical School, Worcester, Massachusetts.
1980 - 1987: Research Assistant in Graphics and Computer Vision at The University of North Carolina at Chapel Hill.
Publications
Lawe, D., A. Chawla, E. Merithew, J. Dumas, W. Carrington, K. Fogarty, L. Lifshitz, R. Tuft, D.Lambright, and S. Corvera," Sequential roles for phosphatidylinositol 3-phosphate and Rab5 in tethering and fusion of early endosomes via their interaction with EEA1",J. Biol. Chem., 277(10), March 8, 2002, pp. 8611-8617.
I performed an analysis of the size of vesicles in cells containing either EEA1 wildtype or EEA1 mutant. By segmenting out the vesicles and calculating sizes I was able to show that EEA1 wildtype tends to produce smaller vesicles than EEA1 mutant. This helps bolster the argument that functioning EEA1 is important in tethering and fusion of early endosomes.
Koji M., T. Yano, D. Schmidt, T. Tokui, M. Shibata, L. Lifshitz, S. Kimura, R. Tuft, and M. Ikebe ,"Rho dependent agonist induced spatio-temporal change in myosin phosphorylation in smooth muscle cells", J. Biol. Chem, 277(1), Jan. 4, 2002, pp. 725-734.
A time series of 3D images of a fluorescently labeled cell (GFP-RhoA) were produced. I analyzed the changes in this distribution (relative to its distance from the plasma membrane), in response to carbachol stimulation, over time. This showed that RhoA preferentially accumulated near the plasma membrane. These kinetics were consistent with that of sustained myosin phosphorylation, (as visualized with labeled phosphorylated MLC) suggesting involvement of the RhoA pathway in this process.
Stachelek, S., R. Tuft, L. Lifschitz [sp] D. Leonard, A. Farwell, and J. Leonard, "Real-time Visualization of Processive, Myosin 5a-mediated Vesicle Movement in Living Astrocytes", J. Biol. Chem., 276(38), Sept. 21, 2001, pp.35652-35659. ref .
In this paper we show that recycling endosomes in astrocytes exhibit hormone-regulated, actin fiber-dependent delivery to the endosomal resorting pool (from the plasma membrane). Recycling vesicle trafficking (via a myosin 5a motor protein) was imaged in 3D images of cells containing GFP labeled vesicles. Images were acquired in 200msec every 15 seconds for 10 minutes. DAVE's 4D visualization was used to appreciate the vesicle movement. Using DAVE's 3D "intelligent cursor", vesicles were manually tracked in 3D over time. I then wrote software to compute vesicle velocities along the actin fibers (automatically skipping periods when vesicles detached). Velocities were ~100nm/sec in the presence of the hormone T4. This is the first demonstration of processive hormone-dependent myosin 5a movement in living cells.
Kirber, M.T., E.F. Etter, K.A. Bellvé, L.M. Lifshitz, R.A. Tuft, F.S. Fay, J.V. Walsh, Jr., and K.E. Fogarty. "The Relationship of Ca2+ Sparks to STOCs Studied with 2D and 3D Imaging in Feline Oesophageal Smooth Muscle Cells". Journal of Physiology, vol. 531(pt. 2), March, 2001, pp. 315-327.
In this paper I used a spatial clustering measure to quantify the 2D clustering of sparks observed in a cell over time. By comparing this to results from a Monte Carlo simulation (and after correction for edge effects due to the shape of the cell), statistical significance of this clustering was shown. Since this simulation assumed equal volume above the entire 2D projection of the cell (i.e. equal probability that any 2D pixel possessed a spark site) , another test of spatial clustering was also designed and applied. It was restricted to testing clustering near the (2D silohouette of the) plasma membrane in a direction tangential to the membrane, thus minimizing 3D volume effects; it too showed significant clustering. Finally, I designed a test of the hypothesis that sparks without STOCS were a different distance from the plasma membrane than sparks with STOCS. This was applied to 3D data, which very poorly defined the location of the top and bottom surfaces of the cell (along the optical axis of the microscope); so distance in one direction was uncertain, while distance within the image plane was known fairly accurately. No significant difference in distance was found.
Patki, V., J. Buxton, A. Chawla, L. Lifshitz, K. Fogarty, W. Carrington, R.Tuft, and S. Corvera. "Insulin action on GLUT4 traffic visualized in single 3T3-L1 adipocytes using ultra-fast microscopy", Molecular Biology of the Cell, vol. 12(1), January, 2001, pp. 129-141. ref
DAVE's 4D visualization (3D over time) was used to understand the motion of GLUT4 vesicles over time. Maximum intensity projection of each 3D image (to produce a 2D time series) was also used (especially for the images in the paper).
ZhuGe, R., K.E. Fogarty, R.A. Tuft, L.M. Lifshitz, K.Sayer and J.V. Walsh, Jr. "Dynamics of Signaling between Ca2+ Sparks and Ca2+ activated K+ Channels Studied with a Novel Image-based Method for Direct Intracellular Measurement of Ryanodie Receptor Ca2+ Current", Journal of General Physiology, vol. 116, 2000, pp 845-864.
Sparks near the cell plasma membrane release calcium which triggers a K+ current (STOC) through BK channels in the membrane. Using reaction-diffusion simulation software which I've developed, I simulated the observed fluorescence in a labeled cell in response to constant Ca2+ current influx. This showed the total fluorescence change is proportional to current over physiological ranges. Simulations also showed that Ca2+ concentration very near a Spark rises very quickly to a high level, suggesting that channel dynamics are due to their own kinetics and not variation in Ca2+ triggering. I also helped in the extensive statistical analysis of the STOC and Spark data (e.g., formulating testable hypotheses about STOC triggering rates, applying t-tests, etc).
Zou, H., L. Lifshitz, R. Tuft, K. Fogarty, and J. Singer, “Imaging Ca2+ Entering the Cytoplasm through a Single Opening of a Plasma Membrane Cation Channel”, Journal of General Physiology, vol. 114, Oct. 1999, pp. 1-14. Cover photo produced with DAVE.
Analysis of 2D time series of fluo-3 labeled cells in conjunction with current recording captured Ca2+ influx through single caffeine activated cation channels. Observed calcium within the cell (i.e. calcium bound to fluo-3, CaFl) varies due to many factors. Some of them are: competing buffers which bind the Ca2+, Ca2+ current through the channel, and whether the channel is in focus or out of focus (the wide-field microscope has a large depth of field and can capture out of focus signal, unlike a confocal microscope). The equations which model Ca2+ influx and binding are reaction-diffusion equations. These are a set of coupled non-linear, time dependent, partial differential equations. Closed-form solution of these equations is not possible except in a few limiting cases. I therefore wrote a program to simulate current entry and reaction-diffusion in a cylindrical cell (a "smooth muscle cell"). Using cylindrical symmetry I was able to simulate the 3D in 2D (length and radius), significantly speeding the simulation. Arbitrary numbers of buffers with varying kinetics were permitted. Channel Ca2+ influx could either be analytically modeled or taken directly from experimental current traces. Tens of simulations were performed with various concentrations of buffers, varying kinetics and currents, so that sensitivity to these parameters could be determined. Simulation results were converted to 3D, blurred by the point spread function of the microscope, and individual 2D slices were then extracted (i.e., either an in-focus slice in the plane of the channel, or an out of focus slice away from the channel's plane). This permitted us to understand the real world impact of the imaging system on the signal we would detect (e.g., out of focus signal might be misinterpreted as smaller channel current). We were also able to show that the time course of CaFl was consistent with single channel, constant Ca2+ influx.
Meininger, G., E. Moore, D. Schmidt, L. Lifshitz, and F. Fay, “Distribution of Active Protein Kinase C in Smooth Muscle”, Biophysical Journal, vol. 77, May, 1999, pp. 973-984.
The deformable model and matched filter (Lifshitz, 1998 below) were applied to activated PKC/Vinculin dual labeled cells. Colocalization of activated PKC and Vinculin along the plasma membrane was calculated; since uncertainty in the signal position along the optical axis still existed, the definition of "colocalized" had to be extended to include fluorescence at nearby voxels. Cells were stimulated with carbachol or PMA to determine if the distribution of activated PKC within the cell changed. To quantitatively answer this question, the percentage of fluorescence as a function of distance from the plasma membrane (i.e., the deformed surface) was calculated prior and post stimulation. Results indicated that after stimulation activated PKC is no longer membrane confined, but is present throughout the cytoplasm. DAVE was used to visualize the dual labeled images and the changes in the PKC distribution.
Kiefer, C., J. McKenney, J. Trainor, R. Lambrecht, H. Bonkovsky, L. Lifshitz, C. Valeri, and L. Snyder, “Porphyrin Loading of Lipofuscin Granules in Inflammed Striated Muscle”, American Journal of Pathology, Vol. 153(3), September, 1998. ref
Quantitative analysis of the ratio of red autofluorescence of porphyrin to the yellow autofluorescence of the sarcolemmal lipofuscin granules (in striated muscle tissue) in which it resided was performed. Porphyrin was found to be significantly higher in tissue with acute inflammation than tissue with chronic inflammation or no inflammation. This suggests that porphyrin loaded vesicles may form within an acute oxidative context. Associated myoglobin data supports the hypothesis that heme iron-catalyzed generation of hydroxyl radicals may play a role in the process. The spatial distribution of porphyrin within the lipofuscin granules was also examined.
Rizzuto, R., P. Pinton, W. Carrington, F. Fay, K. Fogarty, L. Lifshitz, R. Tuft, and T. Pozzan,”Close Contacts with the Endoplasmic Reticulum as Determinants of Mitochondrial Ca2+ Responses”, Science, Vol. 280, June 12, 1998, pp. 1763-1766.
Because mitochondria might respond to microdomains of high Ca2+ that were generated in their proximity by the opening of IP3-gated channels (in the ER) we conducted high-resolution imaging of the mitochondria and of their relation with the intracellular Ca2+ store (the ER). DAVE was used to visualize the 4D series (3D+time) of moving mitochondria in HeLa cells. It was also used to observe the intertwining of mitochondria and ER (expressing mtGFP and erGFP respectively). In addition, by identifying the surfaces of the mitochondria and the ER, and examining the number of voxels which appeared to contain both signals (i.e., in which mitochondria and ER were in very close apposition). I determined that 5-20% of the mitochondrial surface was close to the ER. These microdomains could allow the rapid uptake of a large amount of Ca2+ by mitochondria, with subsequent rapid diffusion of Ca2+ within the mitochondrial network (as visualized via FRAP experiments).
Lifshitz, L. M., “Determining Data Independence on a Digitized Membrane in Three Dimensions”, IEEE Transactions on Medical Imaging, Vol. 17(2), April 1998, pp. 299-303.
This paper describes methods developed to study the spatial distribution of Protein Kinase C (PKC, a facilitator of molecular motors) and Vinculin (a filament anchoring protein) along the plasma membrane. A deformable model (Lifshitz et al., 1994, below) was used to restrict analysis to the plasma membrane. The maximal response of a matched filter (based upon the spatial intensity distribution of a restored bead) converted flourescent blobs into individual voxels. I then extended the K function test for complete spatial randomness (CSR) to apply to voxelized membranes in 3D (standard theory applies to points on a flat plane). Curvature of the membrane, the discrete nature of the voxelized surface, edge effects, and the large number of points (making direct Monte Carlo simulation based on all the points infeasible) all contributed to a challenging problem. In addition, I developed a method to determine whether the two distributions were independent of each other; this extended flat 2D Monte Carlo methods to random translations of the data sets along the plasma membrane. Results showed that PKC and Vinculin were both non-randomly distributed and not randomly distributed relative to each other.
Bassell G., H. Zhang, A. Byrd, A. Femino, R. Singer, K. Taneja, L. Lifshitz, I. Herman, and K. Kosik,”Sorting of beta-Actin mRNA and Protein to Neurites and Growth Cones in Culture”, Journal of Neuroscience, Vol. 18(1), January, 1998, pp.251-265.
I developed a statistical test determine if Rhodamine labeled B-actin mRNA preferentially located near fluorescein labeled tubulin (microtubules). The mRNA was point-like and the micotubules line-like. The Monte-Carlo test I designed compares the histogram of observed distances from mRNA to microtubule with randomly generated histograms. The analysis was performed over a wide range of conditions to control for experimental uncertainties (e.g., exact location of the plasma membrane). It was shown that B-actin mRNA, to 99% confidence, preferentially locates near the microbules in cultured cerebrocortical neurons.
Fay, F. S., K. L. Taneja, S. Shenoy, L. M. Lifshitz, and R. H. Singer, “Quantitative digital analysis of diffuse and concentrated distributions of nascent transcripts, SC35 and Poly(A)”, Experimental Cell Research (with cover photo), Vol. 231(1), February 25, 1997, pp. 27-37.
The spatial relationship between a newly synthesized RNA (BrUTP) and either an RNA splicing factor (SC35) or older RNA (polyA) was examined. 3D images of fixed dual labeled cells at either short times after transcription of BrUTP (<10 min) or longer times (>40 min) were obtained. Internal background noise levels were determined for each cell by examining regions of the cell from which each signal was known to be excluded. This revealed that significant true signal (~70%) of the SC35 was represented as low level diffuse signal rather than the (2x) brighter "speckles" typically analyzed. Percent colocalization of BrUTP with SC35 and with polyA was calculated. The statistical significance of this colocalization was calculated by ranking it relative to colocalizations obtained via random translations of the two data sets (thus preserving any spatial correlations within each image). Despite low levels of BrUTP colocalization (typically 10-30%), observed overlap was still significantly greater than random with respect to both dim (diffuse) SC35 and bright (speckles) SC35. Similarly BrUTP was calculated to have low colocalization with polyA (<10%), but this was less than random at early time points and greater than random at later time points. I also located specific sites of BrUTP transcription as the positions of maximal response to a matched filter. Colocalization of these sites with SC35 was calculated and statistical significance calculated via comparison with colocalization from 100 Monte Carlo simulations with the sites located randomly within the nucleolus. The fact that a high percentage of BrUTP incorporation did not colocalize with concentrated areas of polyA or SC35 would support the conclusion that processing of RNA occurred throughout the nucleoplasm at the site of transcription.
Lifshitz, L. M., J. Collins, E. Moore, J. Gauch, "Computer Vision and Graphics in Fluorescence Microscopy". In: Proceedings of the IEEE Workshop on Biomedical Image Analysis, IEEE Computer Society Press, Los Alamitos, CA, 1994, pp. 166-175.
This paper presents an early version of DAVE. DAVE was unusual in representing volume data (voxels) as tiny polygons. This permitted both surface data (e.g., the plasma membrane) and volume data (e.g., a 3D microscope image of a cell) to be displayed in a unified manner with both taking advantage of the fast hardware for rendering polygons available on SGI computers. This paper also discusses a deformable model we developed (see Lifshitz et all 1992,below).
Moore, E., E. Etter, K. Philipson, W. Carrington, K. Fogarty, L. M. Lifshitz, and F. S. Fay, “Coupling of the Na+/Ca+2 Exchanger to the Na+/K+ Pump and Sarcoplasmic Reticulum in Smooth Muscle”, Nature, Vol. 365, October 14, 1993, pp. 657-660.
This analysis uses the deformable surface model [described above] to locate the plasma membrane. This permitted analysis to ignore nonspecific label in the interior of the cell. Quantitative calculation of colocalization of Na+/K+ pump, Na+/Ca2+ exchanger, vinculin (a marker for contractile filaments), and calsequestrin (marking Ca2+ in SR stores) was performed, compensating for uncertainty in position along the optical axis (lower resolution in that direction). A matched filter was used to identify signal location. Observed colocalization was compared with colocalization along the membrane due to random chance assuming independent spatially random distributions (i.e., a binomial distribution was used to model it). The Na+/K+ pump, Na+/Ca2+ exchanger and calsequestrin all showed high colocalization with each other, while the Na+/K+ pump showed low colocalization with vinculin. This provides evidence for the Na+/Ca2+ exchanger role in removing Ca2+ from the cell (despite its low Ca2+ affinity), since its close proximity to SR Ca2+ release sites might expose it to much higher Ca2+ levels.
Taneja, K. L., L. M. Lifshitz, F. S. Fay, and R. H. Singer, “Poly(A) RNA Codistribution with Microfilaments: Evaluation by In Situ Hybridization and Quantitative Digital Imaging Microscopy”, The Journal of Cell Biology, Vol. 119(5), December, 1992, pp. 1245-1260.
Fluorescent in-situ hybridization was used to label poly(A) RNA inside human fibroblasts. In addition ribosomes, F-actin, or membrane was labeled. Treatment with various substances (to disrupt possible binding between poly(A) RNA and each structure), showed that binding was primarily with actin microfilaments. I derived algorithms to calculate the colocalization in all the dual labeled images. The design was complicated by the fact that the two step labeling procedure did not produce a direct stoichiometric relationship between structure and label (e.g. one actin molecule could be labeled with a variable number of fluorescent molecules). In addition, precise controls were not available to determine noise thresholds. I therefore derived a volume colocalization statistic and applied it over a range of thresholds to examine its sensitivity to noise assumptions. Most importantly I then used a chi-square test to determine the significance of the calculated colocalization (since a certain overlap is expected due to chance). This was one of the earliest applications of quantitation and statistical confidence testing in the field. It was shown that poly(A) RNA colocalized in a nonrandom manner with all three structures.
Lifshitz, L. M.,K. Fogarty, J. M. Gauch, and E. Moore, “Computer Vision and Graphics in Fluorescence Microscopy”, SPIE Proceedings, vol. 1808, Visualization in Biomedical Computing 1992, R. A. Robb (ed.), October, 1992, pp. 521-534.
This paper presents a discussion of several projects current at the time of publication. It discusses a simple, local (and therefore fast) deformable model we developed. The local nature of the model also made it simple to impose constraints on parts of it (e.g., some user specified parts of the surface could be marked "exact" and hence fixed in space while other parts were allowed to deform). This could be used to find the plasma membrane of a 3D image of a flourescently labeled cell. Its versatility also allowed it to track a 2D image over time; the time series was presented as a 3D data set and vertices of the deformable model were restricted to move only in the xy plane (i.e., spatial direction). An early version of DAVE was presented with voxels rendered as triangles and the ability to render the deforming surface color coded for various energy components. A chi-squared test of independence two labels in a dual labeled cell was also presented as a statistically sound method when applied to binary data.
Lifshitz, L. M., “Model-Based Tracking of Deformable Filaments”, SPIE Proceedings, vol. 1609, Model-Based Vision Development and Tools, R. M. Larson and H. N. Nasr (eds.), November, 1991, pp. 185-197.
By using an interpretation tree with an unusual set of cost functions I was able to find and follow (sometimes) several microtubules in a 2D time series of a moving cell. The interpretation tree was used to match fragments of microtubules (identified using a type of edge detector) to a model of the microtubules obtained manually from the first image in the time series. Costs of a match path included the cost of not matching (i.e., leaving a model microtubule unmatched incurred a high cost even if all fragments match their microtubules well).
Lifshitz, L. M. and S. M. Pizer, “A Multiresolution Hierarchical Approach to Image Segmentation Based on Intensity Extrema”, IEEE Transactions on Pattern Analysis and Machine Intelligence. Vol. 12, No. 6, June, 1990, pp. 529-540.
Examination of how isointensity paths merge into intensity extrema, and how extrema and saddle points merge/annihilate as image resolution is descreased determined a (hierarchical) segmentation of 2D CT images of the abdomen. This paper is based upon my dissertation research. I also discovered situations which did not produce the expected hierarchical segmentation and cases which produced maxima as resolution was decreased.
Lifshitz, L. M., F. S. Fay, S. Gilbert, K. Fogarty, and W. Carrington, “Tracking cellular features using motion constraints and global information”, SPIE Proceedings, vol. 1205, Bioimaging and Two Dimensional Spectroscopy, L. C. Smith (ed.), January, 1990, pp. 93-103.
In this paper I describe in detail an algorithm I developed for tracking moving cells in a 2D time series produced via phase contrast microscopy. Phase contrast produces light and dark regions inside the cell, making segmentation difficult. However it also produces a "halo" around outside of the cell. The extremely thin leading lamellipod may have a very weak halo. After locating potential boundary edge pixels (by thresholding the results of a "halo" matched filter) a closed boundary is formed by examining the geometrical and topological relationship among potential boundary pixels. Constraints relating to the previous position are also used.
Fay, F. S., W. Carrington, L. M. Lifshitz and K. Fogarty, “Three- dimensional analysis of molecular distribution in single cells using the digital imaging microscope”, SPIE Proceedings, vol. 1161, New Methods in Microscopy and Low Light Imaging, J. E. Wampler (ed.), August, 1989, pp. 12-23.
The role of microtubules in white blood cell movement (chemotaxis) is analyzed using an array of tools we developed. 2D time series of phase contrast images were analyzed to reveal cell shape changes in response to nocadozole or cholchicine (which disrupt microtubules). A shape polarity measure was developed for this. Since phase contrast produces cell voxels both brighter and darker than the background (depending upon how the phase difference of two paths), after thresholding several stages of dilation followed by erosion were used to locate the cell. Analysis revealed significant shape change in response to the treatments. Static 3D images of anti-tubulin rhodamine labeled cells (to visualize the microtubules) were also obtained. Finally, the microscope was modified to acquire a 2D time series of simultaneous phase contrast (to see the entire cell) and anti-tubulin was obtained so that the dynamic relationship between cell shape and microtubule position could be analyzed. A polar coordinate system was imposed on the cell; the microtubule organizing center was at the origin. Microtubules were manually identified using interactive software. We were able to conclude that just prior to lamellipod formation (at the leading edge of a cell), microtubules came into close proximity with the cell edge. We hypothesized that in some way microtubule entry into that region (cortex) triggered the reaction which formed the lamellipod.
Carrington, W., K. E. Fogarty, F. S. Fay, and L. M. Lifshitz, “Analysis of 3-D Molecular Distribution with the Digital Imaging Microscope”, Proceedings of the 46th Annual Meeting of the Electron Microscopy Soc. of America, San Francisco Press, 1988, pp. 40-41.
Pizer, S. M., J. Gauch, and L. M. Lifshitz, “Interactive 2-D and 3-D Object Definitions in Medical Images Based on Multiresolution Image Descriptions”, SPIE Conference on Medical Imaging, February, 1988.
Lifshitz, L. M., and S. M. Pizer, “A Multiresolution Hierarchical Approach to Image Segmentation based on Intensity Extrema”. In: Information Processing in Medical Imaging, Proceedings of the 10th International Conference, Plenum Press, N.Y., 1988, pp. 107-130.
Pizer, S. M., H. Fuchs, C. Mosher, L. Lifshitz, G. Abram, Ramanathan, Whitney, Rosenman, Staab, Chaney and Sherouse, “3D Shaded Graphics in Radiotherapy and Diagnostic Imaging”, Proceedings of Computer Graphics '86, National Computer Graphics Association, 1986.
Pizer, S. M., J. J. Koenderink, L. M. Lifshitz, L. Helmink and A. D. J. Kaasjager, “An Image Description for Object Definition, Based on Extremal Regions in the Stack”. In: Information Processing in Medical Imaging, Proceedings of the ninth conference, Kluwer Academic Publishers, 1986, pp. 24-37.
Jaques, P., F. DiBianca, S. M. Pizer, F. Kohout, L. M. Lifshitz, and D.Delany, “Quantitative Digital Fluorography - Computer vs Human Estimation of Vascular Stenoses”, Investigative Radiology, January, 1985, pp. 42-52.
I developed an algorithm and implementation which quantified stenosis (given some initial spatial information interactively).
Book Chapters or Sections
Carrington, W., K. Fogarty, R. Rizzuto, L. Lifshitz, and R. Tuft, “High Resolution 3-D Imaging of Living Cells by Image Restoration”, in Imaging Living Cells, R. Rizzuto and C. Fasolato, eds., Springer, Heidelberg, 1998.
Primarily describes image restoration algorithms developed by Dr. Carrington. I helped with some of the visualization of the results and software implementation.
Lifshitz, L. M., “Tracking Cells and Subcellular Features”. In: Advances in Image Analysis, Y. Mahdavieh and R.C. Gonzalez (eds.), SPIE Press, 1992, pp. 218-243.
I report on both my filament tracking algorithm (as described below) and an algorithm for tracking moving cells in a 2D time series produced via phase contrast microscopy (Lifshitz et al. 1990).
Lifshitz, L. M. and S. M. Pizer. “A Multiresolution Hierarchical Approach to Image Segmentation Based on Intensity Extrema”. In: Computer Vision: Advances and Applications, R. Kasturi and R. Jain (eds.), IEEE Computer Society Press, 1991, pp. 606-617.
This describes the main results from my dissertation.
Carrington, W., K. E. Fogarty, L. M. Lifshitz, and F. S. Fay, “Three Dimensional Imaging on Confocal and Wide Field Microscopes”. In: The Handbook of Biological Confocal Microscopy, J. Pawley, (ed.) Plenum Press, N.Y.,1990, pp. 151-161.
I contributed sections to this chapter which described the needs and uses of computer graphics and computer vision (i.e., visualization and analysis) when imaging cells in 3D. This general description was meant for physiologists. Volume vs. surface based rendering, interactive graphics, the need for image segmentation, and human assisted segmentation concepts were presented. Hardware for analysis, display, and storage were described.
Fay, F. S., K. Fogarty, W. Carrington, and L. M. Lifshitz, “Extraction of Information from Biological Systems Using the Digital Imaging Microscope: Analysis of White Blood Cell Chemotaxis”. In: Optical Microscopy for Biology, B. Herman and K. Jacobson (eds.), Wiley-Liss Publishers, N.Y., 1990, pp. 419-435.
I helped develop a method to interactively identify microtubules (via Rhodamine labeled tubulin) and the cell membrane in a 2D time series of moving newt eosinophils (white blood cells). To determine if microtubule location correlated with direction of cell motion and leading lamellipod formation, this spatial information was quantified by examining relative positions of microtubules and membrane in a normalized radial coordinate system. Analysis showed that microtubules come into close proximity to the membrane just prior to lamellipod formation.
Abstracts
Zou, H., L. Lifshitz, M. Kirber, R. Tuft, K. Fogarty, J. Singer, "Visualization of Ca2+ Entry Through Stretch-Activated Cation Channels", Biophysical Journal 80:1 A503, Biophysical Society, 45th Annual Meeting, Boston, Feb. 17-21, 2001.
ZhuGe, R., K. Fogarty, S. Baker, R. Tuft, L. Lifshitz, J. Walsh, “Individual Ca2+ Spark Sites in Smooth Muscle Differ in Number of Ryanodine Receptors and BK Channels ”, Biophysical Journal 80:1 A301, Biophysical Society, 45th Annual Meeting, Boston, Feb. 17-21, 2001.
Patki, V., J. Buxton, A. Chawla, L. Lifshitz, K. Fogarty, W. Carrington, R.Tuft, and S. Corvera. "Insulin action on GLUT4 traffic visualized in single 3T3-L1 adipocytes using ultra-fast microscopy", poster at the American Society for Cell Biology Annual Meeting, Dec 9-13, 2000. San Francisco, CA. The abstract appears in a supplement to Mol. Biol. Cell 11S:127a, Dec 2000.
Carmichael, J., R. Ikebe, J. Saito, N. Saiki, L. Lifshitz, M. Ikebe, “Differences in Actin binding between Calponin and its phosphorylation-site mutants”, Biophysical Society, 44th Annual Meeting, New Orleans, Feb. 12-16, 2000.
Carrington, W., R. Tuft, L. Lifshitz, K. Fogarty, “3-D Fluorescent Imaging: 100 Nanometer Resolution at 500 Frames/Second”, Biophysical Society, 44th Annual Meeting, New Orleans, Feb. 12-16, 2000 (and platform presentation by W. Carrington).
ZhuGe, R., R. Tuft, L. Lifshitz, K. Fogarty, J. Walsh, “Functional Coupling Ratio Between Ryanodine Receptors and BK Channels is Variable, Causing Ca2+ Sparks of the Same Intensity to Elicit STOCS of Different Amplitude”, Biophysical Society, 44th Annual Meeting, New Orleans, Feb. 12-16, 2000 (and platform presentation by K. Fogarty).
Zou, H., L.M. Lifshitz, R.A. Tuft, K.E. Fogarty, and J.J. Singer. “Imaging Ca2+ entry due to a single opening of a plasma membrane cation channel”. Biophysical Journal 76: A465, 1999.
Lifshitz, L., E. Etter, K. Fogarty, K. Bellve, R. Tuft, J. Walsh, and M. Kirber, “Monte Carlo Simulation Indicates Non-Random Spatial Distribution of Ca2+ Sparks in Smooth Muscle”, 1998 Proceedings of the Society of General Physiologists (Sept. Woods Hole meeting), to be published in the Journal of General Physiology. (Also a poster)
Keifer, C., J. McKenney, J. Trainer, L. Lifshitz, S. Krasnicki, T. Schneider, S. Schrier, J. Morrow, C. Valeri, and L. Snyder, ”Hemoglobin-Spectrin Complexes: Distribution within the Erythrocyte Skeleton”, 37th meeting of The American Society of Hematology, Seattle, Washington, 1995.
Moore, E., W. Carrington, K. Fogarty, L. Lifshitz, and F. Fay, "Interpolated Image Deconvolution and Colocalization Analysis", Biophysical Society Conference, 1995.
Ferris, C., Y. Delville, J. Collins, and L. Lifshitz, "Imaging Synaptic Connectivity with Widefield Digital Microscopy", Society for Neuroscience Abstracts, vol. 20, part 2 (24th Annual Neuroscience Conference), p. 1176 (also a poster), 1994.
Gilbert, S., L. Lifshitz, and F. S. Fay, “Feature analysis of phase contrast images of newt eosinophil chemotaxis”,Physiologist, 32:211a. 1989.
Meininger, G. A., L. M. Lifshitz, and F. S. Fay, “Distribution of Protein Kinase C in Smooth Muscle”, ASCB (Cell Biology) Conference, December, 1991.
Posters/Presentations
Lifshitz, L. M., K. Taneja, R. Singer, and F. S. Fay, "Quantitative Evaluation of the Extent of Colocalization between Two Images", Cell Biology Conference, 1992. Voted best poster in topic area.
Lifshitz, L. M., "Tracking Deformable Filaments", Image Processing in Medical Imaging '91, Kent, England, July 7-12, 1991.
Other
Dictenberg,J., W. Zimmerman, C. Sparks, A. Young, C. Vidair, Y. Zheng, W. Carrington, F. Fay, and S. Doxsey, “Pericentrin and Gamma-Tubulin Form a Protein Complex and are Organized into a Novel Lattice at the Centrosome”, Journal of Cell Biology, vol. 141(1), April 6, 1998, pp. 163-174. Cover photo rendered using Data Analysis and Visualization Environment (DAVE) developed by L. Lifshitz.
Femino, A.M., F.S. Fay, K. Fogarty, and R.H. Singer, “Visualization of Single RNA Transcripts in Situ”, Science, April 24, 1998, vol. 280, pp. 585-590. Cover photo (and others) rendered using Data Analysis and Visualization Environment (DAVE) developed by L. Lifshitz.
Cover photo of Experimental Hematology, vol. 26(7), July, 1998. Dual labelled stem cell rendered using DAVE.
Shi, L. M. Lifshitz, and B. Brill, “The Use of Cluster Analysis Difference Imaging in Cancer Diagnosis Using Radiolabeled Monoclonal Antibodies”, Presented at the International Symposium on Nuclear Medicine, in Bejing, China, Oct. 10-14, 1988.
Pizer, S. M., J. M. Gauch, L. M. Lifshitz, and W. R. Oliver, “Image Description via Annihilation of Essential Structures”. Invited paper, Workshop in Computational Vision/Multiresolution Representation of Images, University of Copenhagen, November, 1987.
Affiliations
Member: IEEE, ACM
Ad Hoc Member: NIH Shared Instrumentation Review Panel