**5.1 Evaluation methods**

The following are measurements of evaluation methods or metrics.

### *Blockchain Applications – Transforming Industries, Enhancing Security, and Addressing Ethical...*


**Figure 2.** *ECNN code execution flow.*


**Figure 3.** *Final results of COVID-19 using ECNN approach.*

$$\text{Quality} = \frac{BP + VM}{BP + VP + BM + VM} \tag{1}$$

$$Precision = \frac{BP}{BP + VP} \tag{2}$$

$$\text{Callback} = \frac{BP}{BP + VM} \tag{3}$$

#### **Figure 4.**

*Processor and related resources occupancy of computing device.*

**Figure 5.** *COVID-19 ECNN graph comparing epochs vs. time.*

$$F-measure = \frac{2\text{x}\text{Precision} \propto \text{Callback}}{\text{Precision} + \text{Callback}} \tag{4}$$

**Data Input:** Our experiment was carried over on a database of 282 x-ray images. **Figure 5** demonstrates the time taken to complete iteration of epochs.

Explains the loss ratio with respect to each epochs during execution (**Figure 6**).

Demonstrates the accuracy achieved against each epochs during execution (**Figure 7**).

Demonstrates the loss reduction and accuracy gain of the training model of ECNN with respect to each iteration (**Figure 8**).

Value loss and value accuracy gained of the ECNN model during training (**Figure 9**).

At a glance representation of the comparison among epochs, loss, accuracy of the ECNN model (**Figure 10**).

**Figure 6.** *COVID-19 ECNN graph comparing epochs vs. loss.*

**Figure 7.** *COVID-19 ECNN graph comparing accuracy vs. epoch.*
