A Novel Computer-Aided Diagnosis Scheme on Small Annotated Set: G2C-CAD

<figure class="figure-group responsive"><div class="sub-figure" id="a"><object data="https://static.hindawi.com/articles/bmri/volume-2019/6425963/figures/6425963.fig.003a.svgz" name="6425963.fig.003a" type="image/svg+xml"></object><div class="fig-label">(a) Original sign patches</div></div><div class="sub-figure" id="b"><object data="https://static.hindawi.com/articles/bmri/volume-2019/6425963/figures/6425963.fig.003b.svgz" name="6425963.fig.003b" type="image/svg+xml"></object><div class="fig-label">(b) Patches generated by DCGAN</div></div></figure>

<p>Samples of DCGAN-generated sign patches and real samples. The image on the left shows examples of the seed samples used for training. Most of the generated samples in (b) are realistic; although some of the generated patches can be recognized as being fake (the 5th one in row 3), they do help to boost the CAD system’s performance.</p>

BioMed Research International

fig3

Figure 3

Figure 3: A Novel Computer-Aided Diagnosis Scheme on Small Annotated Set: G2C-CAD 