Scribd
Upload
654 views5 pages

Skin Cancer Detection System Architecture

The document describes a deep learning system architecture for skin cancer detection that includes pre-processing input images, extracting features from the images, training a convolutional neural network (CNN) model using those features, and using the trained CNN model to generate diagnostic results classifying skin lesions as normal skin or skin cancer. The system flows input images through pre-processing, feature extraction, training a deep learning model, and using the trained model to analyze new images and produce diagnostic results. A data flow diagram is shown to illustrate the process of pre-processing input images, extracting features, training and validating a CNN model, and using the model to analyze new image data and

Uploaded by

Melrown Barretto
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
654 views5 pages

Skin Cancer Detection System Architecture

The document describes a deep learning system architecture for skin cancer detection that includes pre-processing input images, extracting features from the images, training a convolutional neural network (CNN) model using those features, and using the trained CNN model to generate diagnostic results classifying skin lesions as normal skin or skin cancer. The system flows input images through pre-processing, feature extraction, training a deep learning model, and using the trained model to analyze new images and produce diagnostic results. A data flow diagram is shown to illustrate the process of pre-processing input images, extracting features, training and validating a CNN model, and using the model to analyze new image data and

Uploaded by

Melrown Barretto
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd

SYSTEM ARCHITECTURE:

PRE-PROCESSING

SMOOTHING ENHANCEMENT SEGMENTATION

DATASET

FEATURE
EXTRACTION

SKIN CANCER NORMAL SKIN

DEEP LEARNING TECHNIQUES CNN

DIAGNOSTIC RESULTS

Fig: The system architecture frameworks of DL.


Level-0
DATA FLOW DIAGRAM

Pre-processing Data
the data Normalisation
Preprocesing
Dataset (Lung of each
Dataset (Skin images
Cancer in a CT-Scan
Cancer images)
images)
Rescaled
Data

Input images of benign and


malignant Feature
Extraction
Level-1
DATA FLOW DIAGRAM

Validate Generate
Split Dataset Model CNN-Model
Save-model (DL)
(Feature
Vector)
Process the
e model (DL)
a lu
v
ted
Training (feature, ic Result
red
label) P Analysis (Test-
Model to
detect skin)
DL-Model
Training Display the output (Accuracy,
Strategy of diagnostic result
Matching the test data with DL
model
User Data Storage Process DL
Select
Image Store (Skin Cancer images) Preprocessing Chose DL
Create training
data and testing set data Parameter
Optimize feature Set in DL
value
Best
Find best features features
Feature
Engineering
Compare the DL
Load DL model with model Generate
parameters Processing of CNN model Generate skin image data
features, lebel performance
Input Provide data in performance and store the
Result of Skin cancer
required format Standard
detection and
value
Output performance strategy of
diagnosis accuracy)
Skin Cancer images
data
Admin User
Preprocess data

Prepare data for


training

Design and construct


Model

Train DL-Model CNN

Validate and Save


Model

Test Model(new-data)

Result Analysis
(Performance strategy
of diagnosis patient)

You might also like