Parameters Finalization of Scientific Software on Impact of Markers on Oral Cancer Tissue and its comparative cost study with Office Automation software

Prabir Kumar Mitra, Rajni Jindal, T. Madhan Mohan

Abstract

In this paper a detailed case study has been given about finalizing an input format regarding data collection sheet on the study made about impact of molecular markers in oral cancer tissue for both premalignant and malignant. Starting from Primary site of Premalignant and Malignant tissue to treatment details, all the parameters of relevance has been discussed thoroughly with the scientists/doctors of the premier medical institutions and then the format has been finalized. The method is based on the characterization of the software to be developed in terms of project and environment and health related attributes. An approach has also been discussed for the estimation of web enabled application development costs for the same is presented. In this an effort has been made towards calculating the total cost of the scientific software development along with its implementation cost. A graphical presentation was also made to show the cost versus achievement model for the scientific package.

Keywords: Software development effort; Software Cost estimation; parameters considered for Oral cancer Programme, Achievement model

Introduction

The use of computers in enterprises has made a significant change in the working style in the enterprises. Computers are used as an effective management tools by the enterprises which enhanced functionality and productivity of the companies. There is a continuously growing need for special information technology (IT) solutions for enterprises regarding generic core processes, such as order processing, production planning and scheduling, manufacturing,[1 and 2] logistics, or more specific tasks, such as patient-doctor relationship management, etc. Today computers are utilized by the premier medical institutions as an IT tool for managing and treating patient effectively and efficiently. To customize a package for a set of specialized doctor working in a research area and to ask them to focus their research works within some fixed parameters is a commendable job. To convince them to confine their mindset on certain fixed parameters for research work were made possible through brainstorming, discussion forum, etc. It was necessary to bring all the doctors to agree a standardize format for doing the research work. With due support from Department of Biotechnology (DBT) renowned doctors from four premier medical institutions sit together and help us deciding the parameters. For success of any good software, the prerequisite is to have the right parameters on which data are to be collected. Several rounds of discussion took place to evolve the final format for the web enable application.

Any information system is developed based on the system life cycle. It composed of development phases consisting of requirement specification and analysis, system design, coding and testing, etc.[3] Various parallel activities are also executed simultaneously. These are project management, planning and software quality assurance which are equally important in system life cycle. Project planning requires the determination of the necessary resources to complete the project and the development of a schedule to be followed after that. Therefore, the generation of estimates of effort and time is required, a task known as Software Cost Estimation (SCE). The most important cost component in information systems development is the manpower costs related [4] to the development effort [5 and 6] of customized software components. Consequently, SCE is focusing on the prediction of the human effort to specify, analyze, design, code and test software. Other activities such as project management itself, project support and software quality assurance are also taken into account. Effort estimations are helpful both for the IT client and the IT developer. In particular, based on these estimations, the acquiring organization may assess and monitor implementation costs, evaluate bids and develop realistic budgets and schedules. In the rest of the paper we will always use the term "cost" to denote human effort for software development [7].

Input Format for 'Impact of Molecular Markers on Oral Cancer Tissue'

The multicentric project is being carried out at four premier medical institutions to study the evaluation and validations of few known markers in individual patients using standardized method. Evaluation will be carried out on both oral cancer patients samples and premalignant oral lesions with different food habits. Oral cancer with distinct premalignant lesions observed in a majority of Indian patients, needs an understanding of biology of the cancer, to utilize defined molecular markers of clinical significance for better patient management. The markers used are Ki-67, p53 protein and antibodies, Telomerase, EGF-R, H-ras, H-ras-c12,Rb,p16,p21. Molecular analysis has been initiated on the samples and percentage alterations of the biomolecular markers are being recorded, few samples are subjected to quality control at various centers in a double blind manner [7]. A baseline study to assess alterations in osmf cases vis-a vis case control subject has been done in 185 cases at RDCH, Chennai; using p53 expression, proliferation and apoptosis.

The data input format for each patient has been subdivided based on its General Information of the Patient, Patient Habits, Premalignant conditions, Histopathology Reports, Primary Sites, Patient Treatment and Impact of Standardized Markers on each patient. Even at the general Information all renowned doctors have accepted that family History and the relationship with the member having cancer is of great relevance and hence are kept the format. They also agreed and want to know how old is the case and date of its primary treatment started. Apart from this, rest all formal information like age, address, sex are all kept under the general information of the patient.

Inside Patient habits, staring from Tobacco chewing with or without lime, Pan Masala, Gutka and Areca nut Chewing information are to be stored. Scientist believe, not only the the habit but also how frequent and for how long is of more significance and hence it is decided the store this data also in habits information.

Under the Primary Site of occurrence the parameters like Lip, Buccal Mucosa, Tongue and floor of the Mouth are considered. Even within Lip, External Upper Lip and External Lower Lip are considered. For Buccal Mucosa, Mucosa of Upper and Lower Lips, Cheek Mucosa, Retromoral areas, Bucco-Alvcolar Sulci.Similarly for Tongue, Anterior Two-thirds and ventral surface are considered. Scientists are of opinion that these are the primary sites where malignant tissue usually found and have significant contribution over the treatment based on its primary site.

For Premalignant Conditions, parameters like Leukoplakia both homogeneous and non-homogeneous, Melanoplakia, Erythroplakia, Submucous Fibrosis and Lichen Planus are considered.

While finializing the Histopathology data format Scientists/Doctors felt there are two sets of data format needed one for premalignant tissue and the other for malignant tissue. In case of premalignant tissue the Hyperplasia status and dysplasia information is good enough for research study whereas in case of malignant tissue the parameters like tumor size, nodal inflatration, Metasis, TNM pattern and tumor grade all are very important information to be kept.

As per as treatment data is concern the parameters like palliative, Vitamin E, Other antioxidant, Photodynamic Therapy, Laser Excision, Cryo Excision and Exercise are of great relevance for premalignant tissue. In case of Malignant tissue or confirmed case of carcinoma, treatment like Surgery, Radiation, Chemotherapy and combined therapy are of great importance.

Evaluation and validation of impact of Seventeen molecular markers data are needed for the whole exercise. These markers are C13-373, D3S1285, Glut2, D9S1748, D9S171, D9S161, IHC-p53, p53 Antibodies, IHC-p21, IHC-p16, IHC-Rb, IHC-Ki67, IHC-ets, IHC-EFG-R, IHC-H.Ras, Mutation-H Ras and Telomerase activity.

To collect the data from various geographical locations, it is felt that a Web enabled uniform data collection sheet will be designed in the form of a dynamic Web site to receive data from these centres. Based on this, a dynamic Web site is designed and launched.

Software Development Cost Estimation:

A crucial activity in the initial phase is the estimation of the necessary Project development effort and time. This effort and time will play a major role in deciding the the software development cost. There are various ways through which Software Development Cost can be measured like Expert Judgment, Algorithmic Cost Estimation and Estimation by analogy, [8] etc.

Software Cost Estimation(SCE)

From managerial point of view software designing cost or Software Cost Estimation rely mainly of Expert Judgment. This method calculates the man-month effort involved to bring the software into reality or developed the software. This approach results in the generation of estimates in an ad-hoc, non-repeatable and non-controllable manner. The amount of badly estimated software development cost reveals the problems of Expert Judgment system.

A typical algorithmic cost prediction model as stated in [7] for a software development may be of the form :

EFFORT=a(Size)^b

As the Size of the software is increasing means the designing is more complicated and hence the effort requirement is more, so does the cost of the project. Here ‘a’ and ‘b’ are the coefficient of numeric positive values and the values depend on the nature of the software to be developed. In case of standard office automation software the values of ‘a’ and ‘b’ are comparatively less than the values for developing a scientific software like this web enabled application for oral cancer tissues.

Total Software Development Cost = C_uf * Effort + C_tsmm

Here, C_uf -> Cost of Unit Effort

C_tsmm -> Total cost of Scientific Man-month involved for finalizing the format and given to computer professional for development.

Implementation Cost:

In case of Web enabled application, the implementation costs involved, the connectivity of Internet access [9] facilities either through Dial-up or Leased line or may be satellite connectivity. The important thing about connectivity is the reliable high speed connectivity at each center is required and the implementation cost will vary depending upon the type of connectivity each center is having. For the Web application on Oral Cancer Programme, the software along with front end and back end will be made available at the Web server. So, it is the connectivity plus data entry charges [10] from the respective centres will be taken as total implementation cost of the application. Therefore implementation cost for a Web application is comparatively higher than a customized stand alone or LAN based software where connectivity cost will not be there. The Implementation Cost is : C iw = C tc + C h + C de

Where C iw à Implementation Cost for Web Application

C tc à Total cost of Internet Connectivity

C h à Hosting charges for the Web Server

C de à Cost of Data entry

The starting cost of either of the curve are shown represents the cost for finalizing the input format. In case of web application on Impact of Molecular Markers on Oral Cancer Tissue, the cost of finalizing the input parameters calls for a series of discussions with the scientists/doctors and hence the cost is high as compared to the general Office automation software. There is comparatively a less price in coding the design in case of a Web application than complex office automation software. However, the implementation cost of the web application is much higher because it includes the cost of connectivity to Internet.

Achievement Model:

The direct achievement of any software depends on the success of the software based on the purpose for which the software is made. In case of a Office Automation software or general Purpose MIS, the achievement depends [11 and 10] upon how best the software suits and support the management in their decision making purpose. Hence the achievement started just after developing and implementing the software based on users need. Whereas a scientific package [5] like this web application for Oral cancer programme has far reaching achievement. This itself a big achievement of the software to bring all the scientific brain to work together with some common parameters to achieve something like treatment of dreaded disease like cancer. The software facilitates the scientists to discuss, confront and established a model or treatment for the disease seeing each other data. When an area is under research, nobody can comment about the success of the research work but definitely can tell the major breakthrough is expected because of cumulative efforts of the scientists working in the area and the availability of the software tools [4] which will facilitates all scientists to see each others data. Definitely, the achievement like helping the scientists in managing their data and the same will start coming just after implementation of the software. This achievement is more or less the same like general purpose office automation software except the achievement will comes after a delay of time. However, the real or significant achievement will start coming if research activity leads to certain discovery, then its an achievement with huge importance and can not be compared with any other in this case since the work is to find a treatment of the dreaded disease like cancer related to oral cavity.

Conclusions

To develop any application requires skill, to finalize the parameters of a scientific software that too if it is in the research stage then it requires lot of patience and determination. It is basically to bring all the scientific brains together and tell them to confine their thought process with stipulated parameters. Hence to design a Web application for a Scientific software is definitely a time consuming and costly solution as compared to develop a generalized Office automation software for which parameters are more or less fixed. However, the achievement level of scientific package is far more than that of an Office Automation software because it is not only supporting the decision makers in their planning and management process but also concise the entire thought process of many scientists together and it also helps in a research activity, hence the achievement of Scientific Software is manifolds.

Acknowledgements

We would like to thank the Scientists/Doctors from renowned medical Institutions for finalizing the input formats. Thanks to our Seniors in NIC and DBT for reviewing the papers and helping us to improve the paper in various ways.

References

1. R. Giachetti, Standard manufacturing information model to support design for manufacturing in virtual enterprises. Journal of Intelligent Manufacturing 10 1 (1999), pp. 49–60.
2. C. Ptak, E. Schragenheim, ERP, Tools, Techniques, and Applications for Integrating the Supply Chain, The St. Lucie Press, 2000.
3. S. Pfleeger, Software Engineering, Theory and Practice, Prentice-Hall, Upper Saddle River, NJ, 2001.
4. L. Angelis and I. Stamelos, A simulation tool for efficient analogy based cost estimation. Empirical Software Engineering 5 1 (2000), pp. 35–68.
5. An integrated software for Quality(ISO 9001), Environmental(ISO 14001) and Occupational Health and Safety(OHSAS 18001) management. This section is about the different modules. Training and competence. Manager. Software parameters. Web. Objectives and Functions. Allows software administrators to authorize or refuse access to www.isovision.ca/english/01_isovision/module/parametre.html
6. Software Cost Estimation: Metrics and Models. Kim Johnson. Department of Computer Science. University of Calgary. Alberta, CANADA T2N 1N4. johnsonk@cpsc.ucalgary.ca. Abstract .Software cost estimation is the process of predicting the amount of effort required to build a software system sern.ucalgary.ca/courses/seng/621/W98/johnsonk/cost.htm
7. Estimating the development cost of custom software  ARTICLE Information & Management, Volume 40, Issue 8, September 2003, Pages 729-741Ioannis Stamelos, Lefteris Angelis, Maurizio Morisio, Evaggelos Sakellaris and George L. Bleris
8. B. Boehm, Software Engineering Economics, Prentice-Hall, Englewood Cliffs, NJ, 1981.
9. Enhanced weighted round robin schedulers for accurate bandwidth distribution in packet networks • ARTICLE Computer Networks, Volume 37, Issue 5, 22 November 2001,Pages 561-578 A. Francini, F. M. Chiussi, R. T. Clancy, K. D. Drucker and N. E. Idirene
10. R. Jeffery, M. Ruhe and I. Wieczorek, A comparative study of two software development cost modelling techniques using multi-organisational and company-specific data. Information and Software Technology 42 (2000), pp. 1009–1016
11. The impact of organizational commitment, senior management involvement, and team involvement on strategic information systems planning  ARTICLE Information & Management, Volume 39, Issue 6, May 2002, Pages 513-524 Vedabrata Basu, Edward Hartono, Albert L. Lederer and Vijay Sethi