# Cancerfree Intern Journal Club ### --revised 2021![](https://i.imgur.com/XenNLVp.png) ---- ### Journal Framework ![](https://i.imgur.com/CPGYO5k.png) ---- ### prerequisite * [what is neural network](https://www.youtube.com/watch?v=aircAruvnKk) * [Deep through CNN](https://medium.com/@RaghavPrabhu/understanding-of-convolutional-neural-network-cnn-deep-learning-99760835f148) ### Techniques(Based on CT scan images) * Data gathering--->==Deep Learning== * Data processing * Image data augmentation (IADLT) --->increase sample sizes of data --->increase the accuracy of training ![](https://i.imgur.com/aow9OIh.png) * Bat algorithm for data equalization (IBA)--->(1)solve overfitting (2)reduce model noise ![](https://i.imgur.com/wC8yGFL.png) * Object Detection ![image alt](https://miro.medium.com/max/880/1*Mj8WKVKf_RpiAsX3SC1ZdQ.png) * CNN(RCNN ) ![](https://i.imgur.com/3ORS7iN.png) ![](https://i.imgur.com/Ryf5k4G.png) ![](https://i.imgur.com/oEPwWsm.png) ![](https://i.imgur.com/FYhTxdp.png) ![](https://i.imgur.com/BMyxZEs.png) * Intersect Over Union (IOU) * **IOU** -> $\frac{Intersection Area}{Union Area}$ $0 \le IOU\le 1$ ![image alt](https://miro.medium.com/max/578/1*89mVRI2rSbyjpkvujNnwxQ.jpeg) * Region of Interest(ROI) ![](https://i.imgur.com/DvUFUFD.png) * Data exchange(security)--->==Blockchain(distributed ledger)== ![](https://i.imgur.com/tdhoq13.png) * Interplanetary file system(IPFS) ![](https://i.imgur.com/TeGsXdj.png) ![](https://i.imgur.com/KTGVu0w.png) ![](https://i.imgur.com/PNrbsaz.png) ![](https://i.imgur.com/vlMxJ6M.png) ![](https://i.imgur.com/BW86LnS.png =x180) * implement smart contract (solidity) ![](https://i.imgur.com/gvx1jyc.png =x350) ![](https://i.imgur.com/KGvAkPx.png =x350) * Deployed smart contracts ![](https://i.imgur.com/7E5nk3k.png =x350) * Transaction details ![](https://i.imgur.com/36RaIlp.png =x350) * Transaction link on rinkeby etherscan ![](https://i.imgur.com/ZaaqBN5.png =x80) * Transaction deployed successfully ![](https://i.imgur.com/nRrMkSz.png) ---- ### Experimental Results Evaluation #### evaluation explanation * training set * testing set * batch size--->samples trained one time * epoch --->all of the traiing set trained at a time * iteration--->batch size trained at a time * ex:training sets=1000,batch size=10,iteration=100,epoch=1 ![](https://i.imgur.com/oezHiQj.png) ![](https://i.imgur.com/FYUa1da.png) ![](https://i.imgur.com/CBrOmdg.png) ![](https://i.imgur.com/yKVoJsM.png) ![](https://i.imgur.com/Kj3v9qq.png) ![](https://i.imgur.com/UFHNmYl.png)