Walker LandWalker Land

Professor, Department of Bioengineering


607-777-4880

wland@binghamton.edu

 

 

 

Positions and Honors

Positions and Employment

1950 - 1954: United States Air Force (Honorable Discharge)

1958 - 1961: Engineer/ Mathematician; Melpar, Inc.

1961: Mathematician; Scope, Inc.

1961 - 1990: Advisory Systems Engineer; Federal Systems Division, IBM Corp.; responsible for research and development in the following areas: statistical and stochastic processing, Bayesian Inferencing, artificial intelligence, expert systems, coherent processing, neural networks, and guidance and location systems; participated in development and evaluation of guidance systems for the Saturn and Apollo vehicles and worked with post Apollo space configurations, including the Space Shuttle.

1988 - present: Research Professor, Lecturer, Adjunct Lecturer; The Watson School of Engineering and Applied Science, Binghamton University, Binghamton, NY.

1997 - present: Director & Principal Investigator; Computational Intelligence Research Group,The Watson School of Engineering and Applied Science, Binghamton University, Binghamton, NY.

1971 - 1976: Navigation System/Subsystem Design

1977 - 1978: Co-PI; Precision Emitter Location System Design

1978 - 1981: PI; Target Fusion System Design

1981 - 1984: PI; Tactical intelligence System Design

1981 - 1987: Member and organizer of Several Committees, Panels and Workshops in the Military Operational Research Society (MORS)

1985 - 1990 PI; Automated Post-processing Portable Workstation design

1994: Co-PI; PC Based Workstation Design and Static Neural Network Technology Development

1995 - 1996: Chair and Organizer; 1st and 2nd annual workshops in Computational Intelligence, Binghamton University, Binghamton, NY.

1997: Chair and Organizer; invited session; Computational Intelligence: Neural Networks; 1997 IEEE International Conference on Systems, Man, and Cybernetics, October 1997; Orlando, FL.

1998: Tract Chair; “Computational Intelligence and Neural Networks”, 1998 IEEE Systems, Man and Cybernetics Society International Conference; San Diego, CA.

2000 - 2001: Organized Teaching Workshops in Evolutionary Computation and Algorithms; International Congress of Evolutionary Computing; San Diego, CA.

2003: Conference Chair and Organizer; IEEE International Workshop on Soft Computing in Industrial Applications; June 23-25, 2003, Binghamton University, Binghamton, NY.

Selected (from over 200) peer-reviewed research publications (in chronological order)

1. Land, W.H. Jr. and Albertelli, L., “Breast cancer Screening Using Evolved Neural Networks”, Invited paper, 1998 International SMC conference, San Diego, Calif. And published in the conference proceedings.

2. Land, W.H. Jr. and Albertelli, L. “Breast Cancer Screening Using Evolved Neural Networks”, Invited paper, 1998 SMC International Conf. pp. 1619-1624.

3. Land, W.H. Jr. and Albertelli, L., Titkov, Y., Kaltsastis, P. and Seburayno, G., “Evolution of Neural Networks for the Detection of Breast Cancer”, 3rd Intentional IEEE Joint Symposia on Intelligence and Systems, 3rd IEEE Symposium on Intelligence in Neural and Biological Systems, May 21-23, Washington, DC, 1998.

4. Land, W.H. Jr. with Masters T., (consultant) and Maniccam, S., “ An Oracle based on the Generalized Regression

Neural Network” Invited paper for the 1998 IEEE International Conference on Systems, Man, and Cybernetics, (SMC) Oct. 11-14, 1998, San Diego, Calif. (published in conference proceedings), pages 1615- 1618.

5. Land, W.H. Jr. with Morrison, C., Masters, T., Lo J.Y. (of the Duke Univ. Medical center), “ Application of A GRNN Oracle to the Intelligent Combination of Several Benign/Malignant Predictive Paradigms", Invited Paper, presented at ANNIE'99 conference, Nov. 99, St. Louis, published in the conference proceedings.

6. Land, W.H. Jr. with Loren, L, and Masters, T., “Investigation of and Preliminary results for the Inter-Observer Variability Problem using FNA Data”, Invited paper ,International Congress of Evolutionary Computing, July 1999, Washington DC. Paper published in conference proceedings.

7. Land, W.H. Jr. with Lo, J.L., and Morrison, C., “Application of Evolutionary programming and Probabilistic neural networks to Breast cancer Diagnosis”, Presented at the IJCNN conference, July, 1999, held in Washington, DC.

8. Land, W.H. Jr. with Khan, S., and Gheorgihu, M., “Application of Evolutionary Programming and probabilistic neural networks to Estimating the Development of breast Cancer”, 3rd. World Multiconference on Systemics, Cybernetics and Informatics July, 31-Aug. 4 th, 1999, Orlando, Fl. And published in the conference proceedings.

9. Lo, J.L. with Land, W.H. Jr. and Morrison, C., “Evolutionary Programming technique for reducing complexity of artificial neural networks for breast cancer diagnosis,” (2000), Medical Imaging.

11. Land, W.H., Jr., Masters, T., Lo, J.Y., and McKee, D.W., (2001). “Application of Evolutionary Computation and Neural Network Hybrids for Breast Cancer Classification Using Mammogram and History Data”, Proceedings of the 2001 IEEE Congress on Evolutionary Computation, Seoul, Korea, May 27-30, 2001, pp. 1147-1154.

10. Land, W. H., Jr., Masters, T., Lo, J. Y., McKee, D. W., & Anderson, F. (2001). Performance analysis of evolutionary computation (EC)/Adaptive Boosting (AB) hybrids for breast cancer classification. Proceedings of the Combined 4th World Multiconference on Systemics, Cybernetics and Informatics (SCI2001) and the 6th International conference on Information Systems Analysis and Synthesis (ISAS2001).

11. Land, W. H., Masters, T., Lo, J. Y., McKee, D. W., & Anderson, F. R. (2001). New results in breast cancer classification obtained from an evolutionary computation/adaptive boosting hybrid using mammogram and history data. Proceedings of the IEEE Mountain Workshop on Soft Computing in Industrial Applications.

12. Land, W.H., Jr., Sadik, O. A., Leibensperger, D., Breimer, M., “Using Support Vector Machines for Developing and testing Intelligent sensors to Combat Terrorism”, Intelligent Engineering Systems through Artificial neural networks, Vol. 12, pp. 811-816, 2002. Received Best Paper Award.

13. Land, W. H., Jr., Lo, J. A., Velazquez, R., “Using evolutionary programming to configure Support Vector Machines for the diagnosis of Breast Cancer”, Intelligent Engineering Systems through Artificial neural networks, Vol. 12, pp. 249-254, 2002.

14. Land, W. H., Jr., Akanda, A., Lo, J. Y., Anderson, F., & Bryden, M. (2002). Application of support vector machines to computer-aided diagnostics (CAD).Proceedings of the SPIE Medical Imaging Conference.

15. Land, W. H., Jr., Bryden, M., Lo, J. Y., McKee, D. W., & Anderson, F. (2002). Performance tradeoff between evolutionary computation (EC)/ adaptive boosting (AB) hybrid and support vector machine breast cancer classification paradigms. Proceedings of the Annual Congress on Evolutionary Computation (CEC), 2002 World Congress on Computational Intelligence.

16. Land, W., McKee, D., Lo, J., and Anderson, F. (2003). Improving the predictive value of mammography using a specialized evolutionary programming hybrid and fitness functions. Proceedings of SPIE (The International Society for Optical Engineering), Medical Imaging, Vol. 5032, San Diego, CA, February 17-20, 2003.

17. Land, W., McKee, D., Velázquez, R., Wong, L., Lo, J., and Anderson, F. (2003). Application of Support Vector Machines to breast cancer screening using mammogram and clinical history data. Proceedings of SPIE (The International Society for Optical Engineering), Medical Imaging, Vol. 5032, San Diego, CA, February 17-20, 2003.

18. Land, W.H., Jr., Sadik, O.M., Wanekaya, A., Uematsu, M.O., Tiwaah, M., Leibensperger, D., and Wong, L., (2003). ”Using Support Vector Machines for the development and testing of Intelligent Sensors”, Pittsburgh Conference, PITTCON, March, 9-14, 2003, Orlando, FL.

19. Land, W.H., Jr., Sadik, O.A., Leibensperger, D., Wong, L., “Integration of Multi Array sensors and Support vector machines for the detection and classification of organophosphate nerve agents”, SPIE International Aero sense Phonetics for Defense and Security Conference, April, 21-25, 2003, Orlando, FL.

20. Land, W., Anderson, F., Velázquez, R., Gonzalez, R. and Lavin, R.. (2003). Computer Aided Diagnosis (CAD) of Breast Cancer. 2003 IEEE International Workshop on Soft Computing in Industrial Applications; June 23-25, 2003, Binghamton, NY.

21. Land, W.H., Jr., and Verheggen, E., (2003). “Experiments Using an Evolutionary Programmed Neural Network with Adaptive Boosting for Computer Aided Diagnosis of Breast Cancer”, SMCia03 International Workshop, June, 23-25, 2003.

22. Land, W. H., Jr., and Bryden, M., (2003). “Kernel Adatron Implementation fro Breast Cancer Data”, SMCia03 International Workshop, June, 23-25, 2003.

23. Land, W., Wong, L., McKee, D., Masters, T., and Anderson, F. (2003). Breast Cancer Computer Aided Diagnosis (CAD) Using a Recently Developed SVM/GRNN Oracle Hybrid. 2003 IEEE International Conference on Systems, Man, and Cybernetics; October 2-8, 2003, Washington, DC.

24. Land, W., McKee, D., Anderson, F. and Lo, J.. (2004). Breast Cancer Classification Improvements Using a New Kernel Function With Evolutionary-programming-configured Support Vector Machines. The International Society for Optical Engineering (SPIE) Conference on Medical Imaging; February 14-19, 2004, San Diego, CA.

25. Land, W., Wong, L., McKee, D., Masters, T., Anderson, F., Raturi, A. and Lo, Y. (2004). New Results in Computer Aided Diagnosis (CAD) of Breast Cancer Using a Recently Developed SVM/GRNN Oracle Hybrid. Proceedings of The International Society for Optical Engineering (SPIE) Conference on Medical Imaging; February 14-19, 2004, San Diego, CA.

26. Land, W., Wong, L., McKee, D., Masters, T., Anderson, F., and Sarvaiya, S. (2004). Data Fusion of Several Support Vector Machine Breast Cancer Diagnostic Paradigms Using a GRNN Oracle. The International Society for Optical Engineering (SPIE), Defense & Security Symposium; April 12-16, 2004, Orlando, FL.

Ongoing Research Support

EMPIRE, Land (PI): 1/1/02 – 5/31/06

Application of a New Computer Aided Classification (CAC) Prototype to Breast Cancer Screening Using Mammogram and Clinical History Data.

The major goal of the project was to evaluate validity of the EP/AB hybrid's performance on improving the specificity and positive predictive value of mammogram interpretations at 100% sensitivity. Several other paradigms have been developed, implemented, tested and evaluated. Information for writing the final research report is currently being assembled.

Status: Grant recently extended to conduct additional research with the Moffitt Cancer and Research center. This extension been approved to evaluate a recently formulated GRNN oracle, developed by Professor Land and Dr. Timothy Masters.

Role: PI

Ongoing Research Support (with Moffitt cancer and research canter) with funding pending

Statistical Learning and Automated Mammographic Diagnosis

In this work novel automated statistical learning systems will be investigated within the scope of diagnostic mammography. Two automated decision paradigms, one based on support vector machine (SVM) considerations and the other on the Kernel Partial Least Squares (KPLS) framework, will be fully developed and analyzed. These automated learning methods will be compared with the diagnostic capabilities of expert mammographers in classifying breast lesions as benign or malignant. A retrospective case-control design based on receiver operator curve (ROC) analysis will form the basis for the various comparisons, where cases are those women with malignant breast lesions and controls are women with benign findings. Both automated methods may be considered as in intelligent learning systems. In theory both approaches converge to similar decision boundaries. However, the KPLS approach developed in part by Professor Land (Co-Investigator) has many desirable assets relative to its ability to learn and with the ease in determining the system's operating parameters when considering practical real-time applications and real-time machine training. Since the SVM approach is somewhat better understood by those in decision theory research, performance comparisons of both automated systems will be assessed in addition to comparing the machine accuracy with the diagnostic mammography performance of this center based on the traditional radiologist's decision. Essentially, the SVM is acting as a validation metric for the KPLS approach. These decision tools will be developed within the framework of building a high-resolution dual monitor workstation. In developing the automated decision methods, traditional breast cancer risk factors will be fused with measures derived from the image automatically as well as derived by the radiologist's image assessments. Likewise, novel uses of the Gail score will also be investigated within the automated decision modeling process; the inclusion of mammographic density will also be explored. Incorporating risk into he decision process essentially implies that the risk factors responsible for setting the disease in play are also useful for assessing the probability of malignancy. This last point is important when considering that a conditioned risk analysis is imbedded in the study based on a diagnostic (or referral) population.

Status: 2.7 million R01 submitted to NIH and is currently in evaluation. Funding Pending

Tomosynthesis Research Effort

We have several graduate and undergraduate students currently involved in this research, which is being conducted with the Moffitt cancer research center in Florida. The following is some background. Background: Although mammography is currently the most effective method in screening and diagnosing breast cancer, it has limitations due to its inherently projective imaging nature. Breast tomosynthesis is an emerging state-of-the-art three-dimensional (3D) imaging technology that demonstrates significant early promise in screening and diagnosing breast cancer. To avoid the observer interpretation variation and to take the advantages of 3D tomosynthesis, automatic interpretation and diagnosis of breast tomosynthesis is prerequisite for the clinical application of tomosynthesis system. The overall goal is to improve detection and diagnosis accuracy of early breast cancer through application of 3D-based computerized detection and diagnosis (CDD) approach. This research will continue through the summer.

Status: Proposal re- submitted to the ACS at funding level at $45,000.00. Research continuing in registration, feature selection and diagnosis of tomosynthesis images.

Role: Co-PI

Unfunded ongoing research projects

This is an ongoing research project with the Moffitt cancer and research center, which is supporting Dan McKee's research doctoral thesis, Cuong Dang's termination project as well as several other undergraduate students in bioengineering and CS. As a first objective, and using confidential images supplied by the Moffitt cancer center, we have to develop new and accurate segmentation techniques for squamous cell carcinoma, Adenocarcinoma as well as Bronchoalveolar carcinoma, and tested these techniques. We have sent the segmented images to the Moffitt cancer center for independent evaluation. The second major objective is to identify the features, develop and test diagnostic algorithms for the independent and automatic diagnosis of these segmented images. We plan to publish the results of this research as well as write a proposal for funding as soon as we demonstrate feasibility of this new approach. This research will also continue through the summer, and we plan to submit a paper to the international SPIE Medical Imaging symposium which will occur in 2006.

Status: Research ongoing. Paper is currently being prepared. Proposal will be written as soon as segmentation, feature selection and diagnostic approach validated with several lung cancer tissue images.

Role: Co-PI

Other breast cancer Research Efforts

The object of this investigation is to design a low cost, automated mammography interpretation system for the detection and diagnosis of direct and indirect signs of breast cancer, and is the topic of Betsy Verheggen's PhD research. Two distinct systems are designed which will result in an outcome of benign or malignant for the presence of features indicating suspicion of breast cancer. One system is expected to model current radiology practices, where the process by the human observer (radiologist) is modeled using soft computing techniques which reduce the image information by feature extraction to obtain a diagnosis. A second system is designed which eliminates the feature extraction step for detection and diagnosis, relying instead on computer vision of the whole image without direct a priori knowledge extracted by human vision. The proposed CAD methods are expected to target those areas where commercially available CAD systems have not maintained their sensitivity in prospective studies since their introduction to the clinical environment.

Role: Major Advisor

Completed Research Support

National Science Foundation (NSF), Sadik (Co-PI) 4/1/02 - 3/31/03
SGNR: Molecular Design of Intelligent Sensors for Selected Chemical Warfare Agents using Support Vector Machines.

The major goal of the project is to integrate chemical sensing “electronic nose” technology with the Binghamton University Breast Cancer Research Group support vector learning machine technology to detect and identify chemical and biological agents and biological spores.

Role: Co-PI

BAE Systems, Land (PI) 1/1/01 – 12/31/01
Application of Support Vector Machines to Vehicle Health Management.

The major goals of this project were to design and test learning machines for the detection and identification of faults in high performance aircraft.

Role: PI

BAE Systems, Land (PI) 1/1/00 – 12/31/00
Development of New Aircraft Flight Control Computational Intelligence Technologies.

The goal of this project was to design and test new intelligent flight control paradigms for high performance aircraft.

Role: PI

Binghamton University, Land (PI) 1/1/99 – 12/31/99
Development of Breast Cancer Diagnostic Paradigms.

The major goals of this project were to develop evolutionary programming and neural network based breast cancer diagnostic software for fine needle aspirate and mammogram sensor inputs.

Role: PI

Last Updated: 1/15/14