Risk assessment...

I want to show that cycling is no more dangerous than walking or driving.

The problem with that statement is that it is more dangerous than walking or driving.

But the point you need to make/prove with the assessment is that even though the stats show that it is more than twice as dangerous than walking/driving, it still isn't so dangerous that it shouldn't be done.

3860 injuries per billion passenger kilometers or one injury per ~260,000km. Work out how much distance you'll typically cover using a bicycle for inter office travel per year (not commuting). 1000km a year? So you'll have, on average, one injury per 260 years you do this.

Then mention cycle training as a suitable mitigation for further reducing this already low risk, which would only make it safer than you likely having one injury, on average, per 260 years.

Being killed or seriously injured is obviously even far less likely to happen. At 1000km/year the KSI stats work out at, on average, once per 1.5 million years. Banning something for such a low risk is pointless.