Fusarium infection affects plant growth before visible symptoms of the disease. You should measure plant growth rate since your treatment point (plant height, root length, number and area of leaves, stem diameter, plant weight including 2 figures - for roots and for part above the ground). Disease symptoms can be rated by wilt occurrence (% of diseased plants), score of wilt development, length of necroses spreading measured in cut stem, number of diseased leaves per plant, etc., physiologic properties (water content, sugars, chlorophyll amount, Na/K, etc.)
There are a lot of works published about such experiments, one of the most cited is - Duffy, Brion K., and Geneviève Défago. "Zinc improves biocontrol of Fusarium crown and root rot of tomato by Pseudomonas fluorescens and represses the production of pathogen metabolites inhibitory to bacterial antibiotic biosynthesis." Phytopathology 87.12 (1997): 1250-1257.
biocontrol can be effective when u sow the seed treated with biocontrol and in field application through grenching, it can be effective control to control the fusarium wilt of tomato.
In order to determine th disease severity, the best way is to use a scale for the rating based on the severity of infection produced. For example, it can range from 0 to 3.
In your case it could be:
0: completely healthy plants
1: less than 3 leaves wilted
2: from 3-5 leaves wilted
3: plants completely wilted
Of course, the scale also depends at what stage of plant growth you would like to measure the severity.
Another way could be to measure the length of the necrosis due to wilting
I think you must take a rating scale like in the publication suggested by Cassiano forner. But you must also take a fresh weight of aerial part and root (separe the root and aerial part weight).
In addition to plant growth parameters (plant length, fresh or dry weight of the aerial and or root system, etc) and disease severity or disease incidence you can estimate the incubation period (i.e., number of days taken from inoculation to appearance of first disease symptoms), the Standardized area under disease progress curve if you have estimations of disease severity over time, or the intensity of vascular colonization at the end of the experiment.
You can find some examples in the attached papers for Fusarium wilt of chickpeas and Verticillium wilt in olive.
Considering that infected plants will die, the most important variable is wilt incidence, which can also be measured using symptoms on stem cortex, as suggested by Alex Ignatov. On the subject, I suggest you evaluate the occurrence of stem symptoms (necrosis and tissue discoloration) to all plants at the finish of the experiment. Other variables concerning to plant growth also should be continuously evaluated. Intensity of leaf and branch wilt may be used as severity variables, perhaps using disease scales as suggested by Jay Ram Lamichhane. Besides that, consider that biocontrol agents could delay plant infection, so incubation period could become an important epidemiological component for your studies.