AI beats doctors in predicting heart diseases

newscapsule_27_02_19

A research team at Imperial College London and the Medical Research Council has finally been able to use Artificial intelligence to forecast outcomes for heart patients, aiding doctors to find the best treatment for individual patients. The technology outperformed doctors, being able to correctly predict a patient’s diagnosis 75% of the time. It is hoped that the new technology will assist doctors to spot clinically useful information from heart scans that would be missed by a human eye enabling them to take an action at the earlier stage.

Reference link: https://www.bhf.org.uk/what-we-do/news-from-the-bhf/news-archive/2019/february/artificial-intelligence-predicts-outcomes-for-heart-patients

Share

Preparing the Research Design

Once the research problem has been identified, the next task for the researcher is preparing the research design. According to Russell Ackoff, “research design is the process of making decisions before a situation arises in which the decision has to be carried out.” It is the conceptual framework within which the research would be carried out. It is a key aspect as it binds the research project together. Its aim is to provide for the collection of relevant information with minimal expenditure of effort, time and money.

But, whether this can be achieved depends upon a large extent on the research purpose, which is classified into four categories: (i) Exploratory; (ii) Description; (iii) Diagnosis; and (iv) Experimentation. For an exploratory research study, a flexible research design is more appropriate as it provides ample scope for researching various aspects of a problem (E.g.Types of vehicles suitable for the Indian market. This topic provides extensive scope for writing). For a research paper, which requires an accurate description, the research design should be formulated in such a way that, it is unbiased and vouches for the reliability of the collected data and analyzed (E.g.: Percentage of small car segment in Indian market. This topic needs accurate facts and figures).

There are various kinds of research designs, such as, experimental (independent variable is manipulated) and non-experimental (independent variable is not manipulated) hypothesis-testing. Experimental designs can be further grouped into informal and formal. Informal experimental design normally uses a less sophisticated form of analysis. It includes: before and after without control design; after only with control design; before and after with control design. Formal experimental design offers relatively more control and uses precise statistical procedures for analysis. It includes: completely randomized design; randomized block design; Latin square design; and factorial designs.

Important factors to remember while preparing the research design:

  • Objectives of the research study;
  • Means of obtaining the information;
  • Tools for data collection;
  • Data analysis (qualitative and quantitative);
  • Time available for each stage of the research; and
  • Cost involved for the research.

A well-planned research design serves as a blueprint for the researcher even before he actually starts working on his research. This helps him to decide his course of action during various stages of the research, thus saving his time and resources.

Share