Cancer Care Nexus

S. Radha Rani, T. Praveen, G. Srihitha, K. Teja Surya Narayana and N. Harika

Vignan’s Foundation for Science, Technology and Research, Guntur -Tenali Rd, Vadlamudi, Andhra Pradesh 522213,

India

Keywords: Machine Learning, Deep Learning, Convolutional Neural Networks (CNN), Random Forest, Medical

Diagnostics.

Abstract: The Cancer Care Nexus is an AI diagnostics plat- form for early cancer detection of breast, lung, skin, and

blood cancer through machine learning and deep learning algorithms. The conventional diagnosis of cancer

is invasive, time-consuming, and costly, making it inaccessible. The project fills these gaps with the

application of Random Forest algorithms for text diagnosis and Convolutional Neural Networks (CNNs) for

image diagnosis, which are efficient and accurate. Cancer Care Nexus offers a unified and easy-to-use

interface, where health workers can input text or image and obtain reliable predictions, enabling multi- cancer

diagnosis in a single platform. The system is scalable, adaptive, and privacy-friendly, providing secure

processing of medical information. Future enhancement includes the expansion of cancer detection, model

performance enhancement, and real- time predictive analysis integration. The project is a key milestone in

AI-based healthcare making cancer detection faster, more accessible, and more accurate, which translates to

improved patient outcomes in the world.

1 INTRODUCTION

Cancer is still one of the leading reasons for deaths

across the world with millions of fresh cases

diagnosed each year. According to the World Health

Organization (WHO), cancer accounted for nearly 10

million deaths in the year 2020 alone and thus the

urge for early and accurate detection methods is

paramount. Early cancer detection significantly

boosts the chances of survival by providing for timely

and efficacious treatment. However, standard cancer

detection methods, such as biopsies, imaging tests,

and histopathological evaluation, are often invasive,

tedious, costly, and call for specialized infrastructure

available at all times in low-resource settings.

Besides, types of cancers such as breast, lung,

skin, and blood cancer require individual-specific

diagnostic methods due to variance in symptoms, data

structures, and clinical manifestations and thus

rendering an integrated approach for multi-cancer

detection unfruitful.

Cancer Care Nexus solves these problems by

creating a single, AI-based diagnostic platform that

employs machine learning and deep learning

algorithms. This project com- bines several models

to diagnose cancers of different types efficiently and

accurately, providing a scalable and flexible solution

for medical professionals. The system applies

Random Forest models to text-based classification

(breast and lung cancer) and Convolutional Neural

Networks (CNNs) for image-based diagnosis (skin

and blood cancer). By offering a single platform

that supports both text and image inputs, Cancer

Care Nexus makes cancer screening easier and more

accessible to healthcare providers.

Conventional cancer diagnostic approaches are

very much dependent on laboratory tests, imaging,

and biopsies, with some drawbacks. Cancer diagnosis

through biopsies can be several days to weeks long,

hindering treatment decisions, while sophisticated

imaging methods such as MRI and CT scans are

expensive and require high-end equipment and

skilled personnel, thus being out of reach for many in

resource- poor areas. Moreover, tests such as biopsies

and tissue extrac-tions are invasive, are

uncomfortable, and also increase the chances of

complications, with some tests requiring repeated

analyses, further weighing down patients. Cancer

detection is also type-dependent, with breast and lung

cancer depending on patient history and clinical

records, while skin and blood cancer diagnosis

depends on image pattern recognition. Also, most

Rani, S. R., Praveen, T., Srihitha, G., Narayana, K. T. S. and Harika, N.

Cancer Care Nexus.

DOI: 10.5220/0013917900004919

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies (ICRDICCT‘25 2025) - Volume 4, pages

623-630

ISBN: 978-989-758-777-1

623

regions, especially in developing nations, do not have

trained oncologists, radiologists, and pathologists,

leading to hurdles in early diagnosis. AI-based

diagnostic tools are an apparent solution with

automated, effective, and affordable screening,

facilitating bridging of the gap in healthcare

accessibility and early cancer detection.

Cancer Care Nexus circumvents the limitations of

conventional cancer diagnosis by coupling

sophisticated machine learning models into a single-

platform, end-to-end diagnostic solution. For breast

and lung cancer diagnosis, Random Forest models

consume patient medical history, symptoms, and

demographics and yield a non-invasive, affordable,

and scalable solution for early-stage diagnosis. For

skin and blood cancer diagnosis, Convolutional

Neural Networks (CNNs) analyze high-resolution

medical images, including dermatological scans and

microscopic blood smear images, and detect subtle

patterns and abnormalities to improve diagnostic

accuracy over conventional manual inspection. The

system provides a single, user-friendly interface that

integrates both text- and image-based models,

allowing healthcare practitioners to enter patient data

and receive real-time diagnostic predictions. This

reduces the need for multiple independent tools,

accelerating the diagnostic process and enhancing

accessibility, efficiency, and accuracy in cancer

screening and early detection.

The use of Cancer Care Nexus has a number of

benefits, and it is a revolutionary device for cancer

screening. AI- driven models provide high diagnostic

accuracy, improving the chances of early-stage

detection, which raises treatment success rates and

reduces mortality significantly. The platform

combines breast, lung, skin, and blood cancer

screening in one system, eliminating the use of

individual screening devices and making the process

cost-effective and efficient. By reducing dependence

on expensive imaging techniques and specialist

professionals, Cancer Care Nexus enhances

accessibility, particularly in developing nations, and

can be integrated into telemedicine services to

enhance healthcare reach. Its explain- able and

automated results allow non-specialists, including

primary care doctors and community health workers,

to con- duct preliminary screenings. Scalable in

nature, the platform allows for future integration of

new cancer types and improved AI models, ensuring

continuous improvement with medical research

developments. AI-driven automation also speeds up

diagnosis while minimizing operational costs,

making cancer screening cost-effective and

accessible. The system has robust data security

features, complying with HIPAA and GDPR

standards to secure encrypted patient information.

Moreover, its explainable AI models pinpoint the

most significant factors driving predictions, ensuring

transparency, clinical relevance, and trustworthiness,

making Cancer Care Nexus a reliable decision-

support system for doctors.

Khalid et al. suggested a deep learning-powered

breast cancer detection model from computerized

mammograms by utilizing feature selection methods

like low-variance feature elimination, univariate

feature selection, and recursive feature elimination

for improved accuracy. Their research employed a

dataset of 3,002 mammography images from 1,501

participants obtained from February 2007 to May

2015, testing six models of classification random

forest, decision tree, k- nearest neighbors, logistic

regression, support vector classifier, and linear

support vector classifier B. N. Kumar et al. The

outcome indicated high accuracy using less

computational power, and hence the model is

effective in the early detection of breast cancer.

Combining MRI and CNN-based classification V.

Sureshkumaret al, Khalid et al. presented an elastic

solution that optimizes diagnostic processes while

tackling computational issues. Their work adds to AI-

assisted cancer diagnostics, paving the way for en-

hanced predictive analytics and broader dataset

generalization in the future.

Kabiraj et al.suggested a breast cancer risk

prediction model based on ensemble machine

learning methods, namely Random Forest and

Extreme Gradient Boosting (XGBoost). Their

research used a breast cancer dataset of 275 instances

with 12 features to compare the predictive

performance of these algorithms. The findings

exhibited a 74.73% accuracy with Random Forest and

73.63% with XGBoost, indicating the promising

potential of ensemble learning techniques in cancer

risk prediction A. Jafari, et al. Their study is

consistent with several studies focusing on the use of

machine learning for breast cancer detection based on

patient information and risk factors including family

history, physical inactivity, psycho- logical stress,

and differences in breast size. The application of

machine learning in medical diagnosis has been

effective in detecting patterns that might go

unnoticed under conventional techniques, thereby

rendering these models highly useful for the early

detection of diseases. The research adds to the

increasing use of AI-based cancer diagnosis,

driving the construction of stronger predictive models

with a view to improving clinical decision-making

and patient outcomes.

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

624

B. S et al . suggested a machine learning-based

lung cancer detection system to support radiologists

in enhancing diagnostic accuracy and patient survival

rates. The authors investigated several classification

methods, such as Support Vector Machine (SVM), K-

Nearest Neighbor (KNN), Decision Tree, Logistic

Regression, Na¨ıve Bayes, and Random Forest, to

identify lung cancer S. P. Maurya et al. The system

utilized a multi-stage classification method, including

data enhancement and segmentation by thresholding

and marker-controlled watershed techniques R.

Javed,et al. The research proved that machine

learning greatly enhances the detection of lung

cancer, with the Random Forest algorithm having the

best accuracy of 88.5%. The results are consistent

with studies showing the efficiency of AI-based

diagnostic tools in medical imaging, especially in the

detection of early-stage lung cancer P. Chaturvedi et

al. Through the use of machine learning for

automated classification, this method lightens the

workload of radiologists and increases diagnostic

accuracy. The research is an addition to current

development of AI-driven healthcare solutions,

advancing the development of more accurate and

accessible lung cancer detection.

Agarwal et al.suggested a machine learning

approach for the detection of lung cancer with an aim

to minimize human error and maximize diagnostic

accuracy with the help of automation. The research

work used four machines learning algorithms

Random Forest, Logistic Regression, Support Vector

Machine, and Decision Tree S. P. Maurya et al,

applied on a dataset of lung cancer in Google Colab,

which offers a cloud platform that includes GPU

support. The efficiency of these algorithms was

measured across four main parameters: accuracy,

recall, harmonic mean, and precision P. Chaturvedi

et al.. The research emphasized the role of automated

detection systems in reducing diagnostic errors and

improving early detection of cancer. Their results

confirm the findings of ongoing studies showing the

capability of AI-based models in enhancing lung

cancer diagnosis by facilitating quicker and more

accurate screening. The comparative study of several

algorithms helps achieve progress in machine

learning technology in medical diagnosis and assists

in developing cancer detection systems that are

effective and scalable and are fit for medical use.

Shehta et al. suggested a deep learning-based

method for the diagnosis of blood cancer, focusing on

early detection to enhance treatment success rates and

minimize mortality. Their research compared various

deep learning architectures, such as ResNetRS50,

RegNetX016, AlexNet, ConvNext, EfficientNet,

Inception V3, Xception, and VGG19, to determine

the best model for efficient and accurate prediction of

blood cancer. Of these, ResNetRS50 showed higher

accuracy and speed with low error rates, and is a

potential device for early detection of cancer. Their

work aligns with continued attempts to apply deep

learning for medical diagnosis since AI-based models

continue to refine cancer screening through increased

detection precision and less reliance on human

examination. Using deep convolutional neural

networks, their work helps develop computerized and

scalable methods for blood cancer diagnosis that

support the utilization of deep learning in enhancing

clinical outcomes and early intervention plans.

Hemalatha et al. suggested an artificial neural

network (ANN)-based method for the diagnosis of

blood cancer based on sensor-generated

physiological data. Their research em- ployed a

sensor network to record important health parameters,

such as cardiac and respiratory rates, body

temperature, and blood pressure, which were then

classified by an ANN . The model had a 92.1%

diagnostic accuracy, showing that ANN-based

systems can successfully diagnose blood cancer and

learn to perform better as additional data are

introduced. The current research fits within the recent

stream of studies exploring AI-assisted cancer

detection and the contribution of neural networks

toward improving diagnostic accuracy at a reduced

cost and dependency on comprehensive clinical

testing. Through the use of sensor data and automated

classification, this research adds to the creation of

cost-effective, scalable, and real-time solutions for

early detection of blood cancer, providing a potential

alternative for more accessible and faster diagnosis in

medical environments.

Akinrinade et al. suggested a deep learning

approach for skin cancer detection, focusing on early

diagnosis to enhance patient outcomes, especially in

underserved areas. Their research mitigated issues

like class imbalance and dataset constraints by

employing methods such as transfer learning, data

augmentation, and Generative Adversarial Networks

(GANs) to boost model performance. The study used

convolutional neural networks (CNNs) to scan

dermoscopic images and utilized texture-based

features to discriminate between malignant and

benign lesions. It also experimented with sampling

strategies and loss functions to enhance imbalanced

dataset classification accuracy O. Akinrinade et al. It

compared ensemble and hybrid models and identified

the most efficient method of early detection of skin

cancer. Their results are consistent with the recent

progress in AI-based healthcare, illustrating that deep

Cancer Care Nexus

625

learning methods can highly improve skin cancer

diagnosis accuracy and accessibility. By

implementing such models on digital health

platforms, the research helps advance the

development of scalable and automated solutions for

early cancer screening, diminishing reliance on

conventional diagnostic approaches.

Kandhro et al. suggested an advanced deep

learning technique for detecting skin cancer

through upgrading the VGG19 pre-trained model

using max pooling and dense layers. Different pre-

trained models, such as VGG19, ResNet152v2,

InceptionResNetV2, DenseNet201, ResNet50, and

InceptionV3 were employed in their study to derive

features from a skin lesion dataset that included

malignant and benign samples. These features were

then categorized based on machine learning

algorithms including Linear Support Vector Machine

(SVM), k-Nearest Neighbors (KNN), Decision Tree

(DT), and Logistic Regression (LR). The research

proved that the use of the enriched VGG19 (E-

VGG19) model A. Kandhro, e t a l with

conventional classifiers greatly enhanced

classification accuracy. Performance was assessed

through recall, F1 score, precision, sensitivity, and

accuracy, proving the efficacy of hybrid methods in

enhancing skin cancer diagnosis. Their results help

in the continued evolution of AI-driven automated

diagnostic systems, offering clinicians more precise

and effective means of early skin cancer detection,

ultimately enhancing patient outcomes.

2 METHODOLOGY

2.1 Dataset Details

The Cancer Care Nexus system uses a

heterogeneous and well-organized dataset for multi-

cancer detection, including blood, skin, lung, and

breast cancer. Each dataset is designed to fit the

respective machine learning models applied for

detection, providing high accuracy and reliability in

classification. In the case of blood cancer detection,

a dataset of 1,659 cancer and 3,389 normal images

in .bmp format (each 450 × 450 pixels) is utilized as

shown in Figure 1.They is high- resolution

microscopic images that facilitate feature extraction

using deep learning for detecting irregular blood cell

shapes. Likewise, in the skin cancer dataset, there are

569 cancer and 235 normal images in .jpg format

(each 194 × 259 pixels) so that Convolutional

Neural Networks (CNNs) can classify malignant and

benign skin lesions with precision as shown in Figure

For the diagnosis of lung cancer, a structured

dataset of

310 × 16 (.csv format) is employed with

patient data, clinical features, and diagnostic labels.

The breast cancer dataset also has a similar structured

form, 570 × 6 (.csv format), enabling the use of

machine learning classifiers like Random Forest and

Logistic Regression for cancer detection at an early

stage.

This collection of datasets provides a complete

multi-modal analysis, combining text-based and

image-based machine learning models for efficient

and scalable cancer diagnosis to assist healthcare

professionals in making quicker and more accurate

predictions.

Figure 1: Blood cancer.

Figure 2: Skin cancer.

2.1.1 User Interaction & Data Input

The Cancer Care Nexus starts with user interaction,

whereby the healthcare providers choose the cancer

to be diagnosed breast, lung, skin, or blood using a

friendly interface (UI). Based on the choice, the

system calls for text-based input (breast and lung

cancer) or image-based input (skin and blood cancer).

The UI makes sure that the data is captured in the right

format, reminding users if some information is

incomplete or invalid. Valid user input is ensured

through proper validation, and this avoids errors,

allowing high-quality data to enter the system. The

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

626

easy-to-use interface shown in Figure 3 makes

navigation easier, enabling healthcare professionals to

upload patient information and view diagnostic results

in real time. The input is then chan- neled to the

preprocessing module, where data is cleaned and

standardized. This step is essential in minimizing

human error, providing consistency, and facilitating

smooth processing for the following detection

models.

Figure 3: User interface.

2.1.2 Input Preprocessing

After the input data is received, the

preprocessing module cleans and optimizes it

to ensure proper classification. In the case of

text data (lung and breast cancer), the system

performs text normalization by eliminating

inconsistencies like special characters, stop

words, and formatting errors. The system then

tokenizes the text, dividing it into meaningful

parts for machine learning analysis. For visual

data (blood and skin cancer), preprocessing

includes scaling, normalizing, and image

augmentation to improve the performance of the

model. Normalization of images provides

consistency in brightness, contrast, and scale,

while image augmentation operations (including

rotation and flipping) facilitate model

generalization. These preprocessing techniques

remove noise, improve pertinent features, and

ensure the data is in a structured form prior to

input into the detection models. By pre-

processing data, the system reduces human

interaction to a great extent, enhancing

efficiency and ensuring high accuracy in

diagnosis.

2.1.3 Models Used

The Cancer Care Nexus framework integrates

different models, each defined for a specific type of

cancer to offer accurate and rapid diagnosis. A

Convolutional Neural Network (CNN) is utilized for

image classification by the skin cancer detection

module. The model contains three convolutional

layers with 32, 64, and 128 filters (3×3 kernels), each

followed by MaxPooling (2×2) to reduce spatial

dimensions. After feature extraction, the model

flattens data into a 1D vector of size 36,992 that is

input to a dense layer with 512 units and a dropout

layer (0.5) to prevent overfitting. The final sigmoid-

activated dense layer enables binary classification

to produce a total of 18,033,177 trainable parameters.

For blood cancer detection, a more complex CNN

architecture is employed, which consists of six

convolutional layers having successively large filter

sizes (32, 64, 128, and 256 filters). Each

convolutional layer is followed by batch

normalization to normalize training and MaxPooling

(2×2) for dimensionality reduction.

The network comprises fully connected layers of

1,024 and 512 neurons, two dropout layers, and a

sigmoid classification layer for binary output. This

design results in 23,252,929 trainable parameters and

hence makes the model effective to detect intricate

patterns in blood smear images. To detect lung and

breast cancer, random forest classifier is employed

which is designed specifically for dealing with

structured data. Instead of images, these models

operate on numeric and categorical values of data

present in tabular data sets in order to detect patterns

in patients’ history. The Random Forest algorithm

builds numerous decision trees and averages multiple

outputs, increasing classification robustness and

reducing overfitting. By combining CNN-based

image processing for skin and blood cancer and

Random Forest-based structured data processing for

lung and breast cancer, the Cancer Care Nexus system

provides an end-to-end, multi-modal cancer detection

solution. The hybrid approach enhances the

classification’s diagnostic accuracy, providing

uniform classification for various cancers, and

optimizing computational efficiency for real-world

medical use.

2.1.4 Cancer Detection Model Execution

Following preprocessing, the system sends the data

to the corresponding cancer detection model

depending on the chosen type of cancer. In the case of

breast and lung cancer, the system uses a Random

Forest algorithm, which is very effec- tive for

Cancer Care Nexus

627

structured text data analysis in predictive

classification. For skin and blood cancer, the system

uses Convolutional Neural Networks (CNNs), which

are particularly useful in image detection and

detection of abnormalities in medical scans. All the

models were trained on extensive datasets to

identify unique cancer characteristics with high

precision and recall rates. The models process the

input data and output classification results, as shown

in figure 5 forecasting whether the sample is

cancerous or not. The output also comprises

confidence scores, which express the model’s

confidence level in its prediction. Through the use

of sophisticated AI models, Cancer Care Nexus

improves accuracy in diagnosis, lessening reliance

on conventional, time-consuming screening

techniques.

2.1.5 Integration & Result Processing

Figure 4: Proposed architecture.

After the predictions are made by the cancer detection

models, the Integration Layer integrates and

processes these outputs, which is done such that the

data flows smoothly across various cancer detection

modules. The system logically integrates and formats

output, organizing them for clear presentation. It

computes confidence scores to enable healthcare

workers to gauge the validity of the diagnosis. The

integration layer guarantees the detection process to

be efficient and without errors, so that the end results

are free from inconsistencies. It also merges multi-

type cancer diagnoses into a single unified answer as

shown in architecture in figure 4 which is essential for

patients with risk for multiple cancers. The

integration layer is critical to simplify, since

healthcare workers are no longer required to manually

decipher results from more than one system. Rather,

all the diagnostic findings are displayed in one screen

to give a unified and precise medical diagnosis.

Figure 5: Proposed modules.

3 OUTPUT GENERATION &

PRESENTATION

After the results are processed, the output processing

mod- ule formats the diagnostic report for

presentation in the user interface (UI). The results

comprise cancer type classification, probability

scores, and image-based diagnosis supporting

visualizations. To make the information readable, the

module structures and presents the information in a

visual and intuitive fashion, facilitating easier

interpretation by medical specialists. For borderline

predictions, the system can recommend additional

medical assessment, allowing clinicians to make

better-informed decisions. The module also features a

clinical recommendation area, providing suggestions

for the next steps depending on AI analysis. The

output processing guarantees the final results are

trustworthy, properly structured, and readable, aiding

healthcare professionals in providing timely and

accurate diagnoses. This step closes the gap between

AI- derived insights and actual clinical decision-

making.

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

628

Figure 6: Breast cancer predition.

Figure 7: Lung cancer predition.

4 CONCLUSION AND FUTURE

WORK

The Cancer Care Nexus system is an easy-to-use

diagnostic platform combining machine learning

algorithms for detection of breast, lung, skin, and

blood cancers. As shown in figure 6, figure 7, figure

8. It uses random forest classifiers for text cancers and

convolutional neural networks (CNNs) for image

cancers with assured reliability. A common interface

facilitates smooth interaction, with feedback from the

user helping to improve it continuously. Patient data

security features ensure privacy, and the scalability of

the system makes it suitable for real-world

applications. Future developments will involve

cutting-edge deep learning methods, multimodal

input data with text, images, and genomic

information, and subtype-specific models for cancer.

Real-time diagnostic functionality will enhance

processing speed, and personalization through patient

history and genetic information will improve

predictions. Continuous learning will allow models to

adapt, and cross modal data fusion will enhance

analysis. Privacy enhancing methods such as

federated learning will protect patient information,

while international collaborations will increase

datasets, making Cancer Care Nexus a stronger and

wiser diagnostic tool. Figure 9 shows the lung cancer

prediction.

Figure 8: Skin cancer predition.

Figure 9: Lung cancer predition.

Cancer Care Nexus

629

5 RESULTS

Cancer Care Nexus is an artificial intelligence-based

cancer detection platform that is able to detect various

types of can- cers through machine learning

algorithms. The Breast Cancer Module uses a

Random Forest Classifier for text analysis, with an

accuracy of 93%, and the Lung Cancer Module uses

the same algorithm for detecting lung cancer with

an accuracy of 95%. For picture-based detection, the

Skin Cancer Module uses a Convolutional Neural

Network (CNN) to achieve 92% accuracy, while the

Blood Cancer Module is also based on CNN and

gives an 85% accuracy. Through these domain-

specific models being integrated together, Cancer

Care Nexus guarantees an end-to-end effective

diagnostic process and early cancer detection, while

helping healthcare experts to make proper clinical

decisions.

REFERENCES

A. I. Shehta, M. Nasr, and A. E. D. M. El Ghazali,” Blood

cancer prediction model based on deep learning

technique,” Sci. Rep., vol. 15, no. 1, p. 1889, 2025.

A. Jafari, “Machine-learning methods in detecting breast

cancer and related therapeutic issues: a review,”

Computer Methods in Biomechanics and Biomedical

Engineering: Imaging & Visualization, vol. 12, no. 1, p.

2299093, 2024.

B. S, P. R, and A. B,” Lung Cancer Detection using

Machine Learning,” in Proc. Int. Conf. Appl. Artif.

Intell. Comput. (ICAAIC), Salem, India, 2022, pp. 539-

543, doi: 10.1109/ICAAIC53929.2022.9793061.

B. N. Kumar, N. C. Gowda, B. J. Ambika, H. N. Veena, B.

Ben Sujitha, and D. R. Ramani, “An efficient breast

cancer detection using machine learning classification

models,” International Journal of Online & Biomedical

Engineering, vol. 20, no. 13, 2024.

I. A. Kandhro, S. Manickam, K. Fatima, M. Uddin, U.

Malik, A. Naz, and A. Dandoush, “Performance

evaluation of E-VGG19 model: Enhancing real-time

skin cancer detection and classification,” Heliyon, vol.

10, no. 10, 2024.

I. A. Kandhro et al.,” Performance evaluation of E-VGG19

model: Enhancing real-time skin cancer detection and

classification,” Heliyon, vol. 10, no. 10, 2024.

K. Hemalatha et al.,” An Enhanced Analysis of Blood

Cancer Prediction Using ANN Sensor-Based Model,”

Eng. Proc., vol. 59, no. 1, p. 65, 2023.

Khalid, A.; Mehmood, A.; Alabrah, A.; Alkhamees, B.F.;

Amin, F.; AlSalman, H.; Choi, G.S. Breast Cancer

Detection and Prevention Using Machine Learning.

Diagnostics 2023, 13, 3113.

M. Adjouadi et al., “Classification of leukemia blood

samples using neural networks,” Annals of Biomedical

Engineering, vol. 38, pp. 1473- 1482, 2010.

M. K. Monika, N. A. Vignesh, C. U. Kumari, M. N. V. S.

S. Kumar, and E. L. Lydia, “Skin cancer detection and

classification using machine learning,” Materials

Today: Proceedings, vol. 33, pp. 4266-4270, 2020.

O. Akinrinade and C. Du,” Skin cancer detection using deep

machine learning techniques,” Intell.-Based Med., vol.

11, p. 100191, 2025.

O. Akinrinade and C. Du, “Skin cancer detection using deep

machine learning techniques,” Intelligence-Based

Medicine, vol. 11, p. 100191, 2025.

P. Chaturvedi, A. Jhamb, M. Vanani, and V. Nemade,

“Prediction and classification of lung cancer using

machine learning techniques,” in IOP Conference

Series: Materials Science and Engineering, vol. 1099,

no. 1, p. 012059, Mar. 2021, IOP Publishing.

R. Raina et al., “A systematic review on acute leukemia

detection using deep learning techniques,” Archives of

Computational Methods in Engineering, vol. 30, no. 1,

pp. 251-270, 2023.

R. Javed, T. Abbas, A. H. Khan, A. Daud, A. Bukhari, and

R. Alharbey, “Deep learning for lungs cancer detection:

a review,” Artificial Intelligence Review, vol. 57, no. 8,

p. 197, 2024.

S. Kabiraj et al.,” Breast cancer risk prediction using

XGBoost and random forest algorithm,” in Proc. 11th

Int. Conf. Comput., Commun. Netw. Technol.

(ICCCNT), Kharagpur, India, Jul. 2020, pp. 1-6, doi:

10.1109/ICCCNT49239.2020.9225433.

S. Agarwal, S. Thakur, and A. Chaudhary,” Prediction of

Lung Can- cer Using Machine Learning Techniques

and their Comparative Anal- ysis,” in Proc. 10th Int.

Conf. Reliability, Infocom Technol. Optim. (Trends

Future Directions) (ICRITO), Noida, India, 2022, pp. 1-

5, doi: 10.1109/ICRITO56286.2022.9965052.

S. P. Maurya, P. S. Sisodia, R. Mishra, and D. P. Singh,

“Performance of machine learning algorithms for lung

cancer prediction: a comparative approach,” Scientific

Reports, vol. 14, no. 1, p. 18562, 2024.

U. K. Lilhore, S. Simaiya, Y. K. Sharma, K. S. Kaswan, K.

B. Rao, V. M. Rao, A. Baliyan, A. Bijalwan, and R.

Alroobaea, “A precise model for skin cancer diagnosis

using hybrid U-Net and improved MobileNet-V3 with

hyperparameters optimization,” Scientific Reports, vol.

14, 2024.

V. Sureshkumar, R. S. N. Prasad, S. Balasubramaniam, D.

Jagannathan, J. Daniel, and S. Dhanasekaran, “Breast

cancer detection and analytics us- ing hybrid CNN and

extreme learning machine,” Journal of Personalized

Medicine, vol. 14, no. 8, p. 792, 2024.

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

630