The American Journal of Medical Sciences and Pharmaceutical Research
27
https://www.theamericanjournals.com/index.php/tajmspr
TYPE
Original Research
PAGE NO.
27-37
10.37547/tajmspr/Volume07Issue01-04
OPEN ACCESS
SUBMITED
16 October 2024
ACCEPTED
09 December 2024
PUBLISHED
10 January 2025
VOLUME
Vol.07 Issue01 2025
CITATION
Han Thi Ngoc Phan, & Arjina Akter. (2025). Predicting the Effectiveness of
Laser Therapy in Periodontal Diseases Using Machine Learning Models.
The American Journal of Medical Sciences and Pharmaceutical Research,
7(01), 27
–
37. https://doi.org/10.37547/tajmspr/Volume07Issue01-04
COPYRIGHT
© 2025 Original content from this work may be used under the terms
of the creative commons attributes 4.0 License.
Predicting the
Effectiveness of Laser
Therapy in Periodontal
Diseases Using Machine
Learning Models
Han Thi Ngoc Phan
1
, Arjina Akter
2
1
Dentist, Pham Hung Dental Center MTV Company Limited, Pham Hung
Street, Binh Chanh district, Ho Chi Minh city, Vietnam
2
Department Of Public Health, Central Michigan University, Mount
Pleasant, Michigan, USA
Abstract:
This study evaluates the effectiveness of
machine learning models in predicting the outcomes of
laser therapy for periodontal diseases. Various
algorithms, including Neural Networks, Gradient
Boosting, Random Forest, and Support Vector Machine,
were applied to a dataset containing clinical variables
such as pocket depth and gingival inflammation. The
Neural Network model achieved the highest predictive
accuracy with an AUC-ROC score of 0.91, followed by
Gradient Boosting at 0.90. These models outperformed
traditional techniques, demonstrating that machine
learning can accurately predict treatment success. The
findings suggest that machine learning can aid clinicians
in personalizing laser therapy, optimizing treatment,
and improving patient outcomes. Further research with
diverse datasets is recommended to refine these
models.
Keywords:
Machine learning, laser therapy, periodontal
diseases, predictive accuracy, AUC-ROC, Neural
Networks, Gradient Boosting, treatment outcomes,
clinical variables, personalized treatment.
INTRODUCTION:
Periodontal diseases, commonly referred to as gum
diseases, are inflammatory conditions affecting the
tissues surrounding and supporting the teeth. Left
untreated, they can lead to tooth loss and systemic
complications. Traditional approaches to treating
periodontal diseases include scaling, root planing, and
surgical interventions. However, recent advancements
have introduced laser therapy as a promising non-
The American Journal of Medical Sciences and Pharmaceutical Research
28
https://www.theamericanjournals.com/index.php/tajmspr
The American Journal of Medical Sciences and Pharmaceutical Research
invasive treatment option. Laser therapy offers the
potential for improved clinical outcomes, including
reduced inflammation, enhanced tissue healing, and
minimized discomfort for patients.
Despite the growing adoption of laser therapy, its
effectiveness in treating periodontal diseases has been
subject to varying opinions within the dental and
medical communities. Machine learning, with its ability
to process vast amounts of data and identify patterns,
provides an innovative approach to evaluating the
clinical effectiveness of laser therapy. By leveraging
machine learning algorithms, we aim to predict
treatment outcomes based on clinical variables and
patient characteristics. This study explores the
application of machine learning models to assess the
effectiveness of laser therapy and provides actionable
insights for clinicians to optimize treatment strategies.
The focus of this research is to not only evaluate the
predictive accuracy of various machine learning
models but also to identify key clinical factors
influencing
treatment
success.
The
results
demonstrate the potential of advanced machine
learning techniques, such as neural networks and
gradient boosting, to enhance decision-making in
periodontal therapy.
LITERATURE REVIEW
The integration of laser therapy into periodontal
treatment has been widely studied over the past
decade. Lasers, such as diode lasers and erbium lasers,
have demonstrated efficacy in reducing bacterial load,
improving tissue healing, and promoting regeneration
of periodontal structures. A study by Schwarz et al.
(2008) highlighted the bactericidal effects of lasers and
their ability to achieve comparable or superior
outcomes to traditional methods. However, the
variability in clinical success across patient populations
underscores the need for personalized treatment
approaches.
Recent advancements in machine learning have
introduced a new paradigm for analyzing treatment
outcomes in healthcare. Machine learning algorithms
have been applied to predict disease progression,
treatment success, and patient-specific risk factors. In
the context of periodontal diseases, machine learning
provides an opportunity to assess large datasets and
extract meaningful insights that guide clinical decision-
making.
Previous studies, such as Lee et al. (2020), applied
machine learning to predict periodontal disease
progression based on clinical and genetic markers.
Their findings demonstrated that models like Random
Forest and Support Vector Machines are effective in
identifying high-risk patients. However, limited studies
have focused specifically on evaluating the outcomes of
laser therapy using machine learning.
This research builds upon the existing literature by
introducing a comprehensive machine learning
framework to evaluate laser therapy outcomes. Our
methodology incorporates clinical variables such as
pocket depth, gingival inflammation, and laser therapy
duration. The results, as outlined in this study, indicate
that neural networks and gradient boosting outperform
traditional methods, achieving an AUC-ROC of 0.91 and
0.90, respectively. These findings are consistent with
prior research, such as Pereira et al. (2019), which
highlighted the potential of neural networks in dental
treatment predictions.
This study advances the literature by not only
confirming the clinical effectiveness of laser therapy but
also demonstrating how machine learning can provide
personalized
treatment
recommendations.
By
integrating predictive modeling with clinical practice,
this research bridges the gap between technological
advancements and evidence-based dentistry, offering a
path toward more effective and patient-centered
periodontal care.
METHODOLOGY
This study aims to assess the effectiveness of laser
therapy in treating periodontal diseases using machine
learning techniques. Periodontal disease, which affects
the tissues surrounding and supporting the teeth, can
lead to severe dental health issues if not treated
properly. Traditional methods of treatment often
involve scaling, root planning, and the use of antibiotics,
but advancements in technology, particularly the
application of laser therapy, have shown promising
results. Laser therapy is considered a more efficient and
less invasive approach, leading to quicker recovery
times and improved patient outcomes. However, there
is limited research quantifying its effectiveness using
advanced computational techniques like machine
learning. This study addresses this gap by applying
machine learning models to a dataset containing clinical
parameters, treatment details, and patient responses to
laser therapy. The ultimate goal is to build a predictive
model that can forecast the success of laser therapy
based on various patient-specific and treatment-related
factors.
The dataset comprises patient records, clinical
measurements, and details on their response to laser
therapy. Various machine learning algorithms are
employed to evaluate how demographic factors,
disease severity, laser therapy settings, and post-
treatment responses can be utilized to predict the
The American Journal of Medical Sciences and Pharmaceutical Research
29
https://www.theamericanjournals.com/index.php/tajmspr
The American Journal of Medical Sciences and Pharmaceutical Research
effectiveness of laser treatment. These predictions will
assist dental professionals in making more informed
decisions regarding treatment plans for periodontal
diseases.
DATASET DESCRIPTION
The dataset used for this study is sourced from clinical
trials, patient treatment histories, and observational
studies conducted at multiple dental clinics and
hospitals specializing in periodontal care. This dataset
spans a range of variables that cover patient
demographics, pre-treatment conditions, the specifics
of laser therapy (such as duration and frequency), and
post-treatment
outcomes.
Each
data
record
corresponds to a unique patient, and the dataset
includes both categorical and continuous variables,
reflecting the complexity of the treatment process and
patient health.
The data collected includes parameters such as the
patient's age, gender, clinical indicators like pocket
depth and bleeding, as well as the specifics of laser
therapy including duration and frequency. Post-
treatment data such as pocket depth reduction,
presence of bleeding, and overall clinical improvement
or worsening are also recorded to evaluate the success
of the therapy. The dataset is designed to include both
before-and-after treatment data, making it suitable for
analyzing the impact of laser therapy on periodontal
disease over time.
A summary of the dataset is presented in Table 1, which
includes details of each feature collected.
Table 1: Dataset Overview
Feature
Description
Type
Patient_ID
Unique identifier for each patient
Categorical
Age
Age of the patient in years
Continuous
Gender
Gender of the patient (Male/Female)
Categorical
Pre_Treatment_Pocket_Depth
Pocket depth (in millimeters) before treatment
Continuous
Post_Treatment_Pocket_Depth
Pocket depth (in millimeters) after treatment
Continuous
Pre_Treatment_Bleeding
Bleeding on probing before treatment (Yes/No)
Categorical
Post_Treatment_Bleeding
Bleeding on probing after treatment (Yes/No)
Categorical
Laser_Therapy_Duration
Duration of laser therapy (minutes)
Continuous
Laser_Therapy_Frequency
Frequency of laser therapy sessions (times/week)
Continuous
Clinical_Outcome
Clinical outcome (Improved/Unchanged/Worsened) Categorical
Gingival_Inflammation
Level of inflammation (mild/moderate/severe)
Categorical
Follow_Up_Period
Time elapsed post-treatment (months)
Continuous
DATA PREPROCESSING
The data preprocessing phase is essential for
transforming the raw dataset into a form that is
suitable for training machine learning models. The first
step involves identifying and addressing missing data
points. Missing data can significantly reduce the
accuracy and reliability of machine learning models.
Therefore, missing values are imputed using
appropriate techniques. Continuous variables with
missing data will undergo mean imputation, where the
missing value is replaced by the mean of the available
data points for that feature. For categorical variables,
missing entries are imputed with the mode (most
frequent value) or, in some cases, the records may be
removed if missing data is substantial.
Next, continuous features, such as age, pocket depth,
and laser therapy duration, are scaled to a standard
range. This is necessary because machine learning
algorithms often perform better when features have
similar scales, ensuring that no single feature
dominates due to its scale. Min-Max scaling will be used
to normalize these features, bringing them to a range
between 0 and 1. This step is critical for algorithms like
Support Vector Machines (SVMs) and neural networks,
which are sensitive to the magnitude of the features.
Categorical variables, such as gender, bleeding before
treatment, and clinical outcomes, will be encoded using
one-hot encoding. This technique converts categorical
variables into a binary vector, where each possible
category is represented by a binary column (1 if the
category is present, 0 otherwise). For example, the
variable "gender" will be encoded into two columns,
one for male and one for female, with a 1 in the
respective column for each patient.
Outlier detection is another key component of
preprocessing. Outliers in continuous features can
disproportionately influence the model's performance
and lead to biased results. Statistical techniques, such
as Z-score or interquartile range (IQR), will be applied
to identify and handle outliers. In some cases, extreme
values will be removed if they are deemed erroneous
or unrepresentative of the population.
The American Journal of Medical Sciences and Pharmaceutical Research
30
https://www.theamericanjournals.com/index.php/tajmspr
The American Journal of Medical Sciences and Pharmaceutical Research
Finally, the dataset will be split into training and testing
sets. Typically, 70% of the data will be used for model
training, while the remaining 30% will be reserved for
testing. This ensures that the model is evaluated on a
separate set of data that it has not encountered during
training, providing an unbiased estimate of its
performance.
Feature Selection
Feature selection is the process of identifying the most
relevant features that contribute to the model's ability
to predict treatment outcomes. This is a crucial step, as
it helps improve the model's performance, reduce
overfitting, and decrease computational complexity.
Various techniques will be used to select the most
important features for the model.
First, a correlation analysis will be performed to
understand the relationships between continuous
features, such as pocket depth, age, and laser therapy
duration. Pearson’s correlation coefficien
t will be
computed to quantify the strength and direction of
linear relationships between these features and the
target variable (clinical outcome). Highly correlated
features may be dropped to avoid redundancy,
ensuring the model does not overfit.
Chi-square tests will be applied to assess the
relationships between categorical variables, such as
pre-treatment bleeding and clinical outcome. This test
evaluates whether the observed distribution of
categorical variables differs significantly from what
would be expected under the assumption of
independence. If the p-value is significant, the feature
will be retained.
Additionally, Recursive Feature Elimination (RFE) will be
employed using a machine learning model, such as
Random Forest, to recursively eliminate less important
features. RFE evaluates the importance of each feature
based on the model’s performance and removes those
that contribute minimally to predicting the outcome.
Machine Learning Models
To evaluate the effectiveness of laser therapy, several
machine learning algorithms will be trained on the
processed dataset. Each model is selected based on its
suitability to the type of data and the problem at hand.
Logistic Regression is chosen as the baseline model due
to its simplicity and interpretability. This model will
predict the likelihood of a binary clinical outcome
(improved or worsened) based on the input features.
Logistic regression is particularly useful when the
relationship between the dependent and independent
variables is expected to be linear.
Random Forest Classifier will be used as an ensemble
learning technique to capture non-linear relationships
and interactions between features. Random forests are
particularly useful for feature importance analysis,
allowing the model to rank features based on their
contribution to predicting treatment outcomes.
Support Vector Machine (SVM) will be employed to
classify the clinical outcomes based on the features.
SVMs are powerful for classification tasks, particularly
when dealing with high-dimensional feature spaces and
small to medium-sized datasets. They are capable of
handling complex decision boundaries.
Gradient Boosting Machines (GBM) will be used to
create an ensemble of decision trees that are built
iteratively to minimize the prediction error. GBMs are
known for their high accuracy and ability to handle both
categorical and continuous data effectively. Neural
Networks will be applied to explore more complex
patterns within the data. Due to their ability to model
non-linear relationships and interactions, deep learning
models like neural networks are highly suitable for large
and high-dimensional datasets.
Each model will be trained and tested on the dataset to
evaluate which algorithm provides the best
performance in predicting the effectiveness of laser
therapy.
Model Evaluation
To assess the performance of the machine learning
models, multiple evaluation metrics will be used.
Accuracy will be computed as the proportion of
correctly predicted instances out of the total
predictions. However, given the potential for
imbalanced classes in the dataset (e.g., a majority of
patients may show improvement), accuracy alone may
not be sufficient.Precision, recall, and F1-score will also
be calculated to provide a more nuanced
understanding of model performance. Precision
measures the proportion of true positives among all
predicted positives, while recall measures the
proportion of true positives among all actual positives.
The F1-score, which is the harmonic mean of precision
and recall, provides a balance
d measure of the model’s
performance, especially in cases of class imbalance.
The AUC-ROC curve will be plotted to evaluate the
model’s ability to discriminate between positive and
negative outcomes. The Area Under the Curve (AUC)
provides a single scalar
value representing the model’s
ability to rank predictions correctly, with a higher AUC
indicating better performance. A confusion matrix will
also be constructed to visualize the model's
performance across the different classes (improved,
worsened, unchanged). This matrix shows the number
of true positives, true negatives, false positives, and
false negatives. Finally, K-fold cross-validation will be
used to assess the model’s performance by training and
The American Journal of Medical Sciences and Pharmaceutical Research
31
https://www.theamericanjournals.com/index.php/tajmspr
The American Journal of Medical Sciences and Pharmaceutical Research
testing the model on different subsets of the data. This
technique helps ensure that the model generalizes well
and is not overfitting to a specific portion of the data.
Hyperparameter Tuning
To optimize the performance of the machine learning
models, hyperparameter tuning will be conducted
using either grid search or randomized search
techniques. Hyperparameters such as learning rate,
regularization strength, number of estimators (in
ensemble models), and tree depth (in decision tree-
based models) will be fine-tuned to maximize the
model's accuracy.
Grid search exhaustively tests all possible combinations
of hyperparameters, while randomized search selects
random combinations, which can be more efficient
when dealing with a large number of hyperparameters.
Interpretability and Model Explainability
Given the complexity of machine learning models,
particularly deep learning and ensemble methods,
model interpretability and explainability will be
prioritized. SHAP (Shapley Additive Explanations) will
be used to explain the contribution of each feature to
individual predictions. SHAP values quantify the impact
of each feature on the model’s output, offering a
detailed understanding of how specific features
influence the predicted clinical outcome.Feature
importance will also be extracted from tree-based
models, such as Random Forest and Gradient Boosting,
to identify the most influential factors driving
treatment outcomes. These insights can be valuable for
clinicians when designing personalized treatment plans
for periodontal disease patients.
Results Interpretation
The results from the trained machine learning models
will be analyzed to assess the effectiveness of laser
therapy. Key factors such as laser therapy duration,
frequency, and pre-treatment conditions like pocket
depth and gingival inflammation will be evaluated to
understand their impact on treatment outcomes. The
final model will provide a set of predictions for new
patient data, aiding in the identification of those who
are likely to respond well to laser therapy and those
who may require alternative treatment options.
Ethical Considerations
Throughout the study, ethical guidelines will be
adhered to, ensuring that patient data is anonymized,
and that patient confidentiality is maintained. Informed
consent will be obtained from all participants whose
data is used in the study, ensuring that they understand
the purpose of the research and how their data will be
used. The study will comply with all relevant ethical and
legal regulations governing the use of medical data.
This methodology outlines a comprehensive approach
to evaluating the effectiveness of laser therapy in
treating periodontal diseases using machine learning
techniques. By analyzing clinical data and leveraging
powerful machine learning models, the study aims to
provide evidence-based insights that can guide clinical
decision-making in periodontal care. Through this
research, we hope to contribute to the growing div of
knowledge
on
the
application
of
advanced
computational methods in healthcare, particularly in
the field of periodontal disease treatment.
RESULT
Overview of Model Performance
The machine learning models were trained and tested
on the dataset described in the methodology section.
The goal was to evaluate the effectiveness of laser
therapy in treating periodontal diseases by predicting
clinical outcomes, such as improvement, lack of
change, or worsening of the disease. Various models,
including Logistic Regression, Random Forest, Support
Vector Machine (SVM), Gradient Boosting, and Neural
Networks, were applied to the dataset, and their
performances were evaluated based on key metrics
such as accuracy, precision, recall, F1-score, and Area
Under the Curve (AUC).
Evaluation Metrics
To assess the predictive power of the models, the
following evaluation metrics were used:
•
Accuracy: The proportion of correct predictions
out of all predictions.
•
Precision: The proportion of true positives
among all predicted positives.
•
Recall: The proportion of true positives among
all actual positives.
•
F1-Score: The harmonic means of precision and
recall, providing a balanced measure of the
model’s performance.
•
AUC-ROC Curve: The area under the receiver
operating characteristic curve, which indicates
the model’s ability to discriminate between the
positive and negative outcomes.
These metrics were used to compare the performance
of the different machine learning models and
determine the most effective approach for predicting
clinical outcomes in periodontal disease treatment
using laser therapy.
The American Journal of Medical Sciences and Pharmaceutical Research
32
https://www.theamericanjournals.com/index.php/tajmspr
The American Journal of Medical Sciences and Pharmaceutical Research
Model Comparison and Results
The models were evaluated based on the metrics, and the following results were observed:
Table 2: Model Performance Comparison
Model
Accuracy Precision Recall F1-Score AUC-ROC
Logistic Regression
0.75
0.72
0.78
0.75
0.82
Random Forest
0.82
0.80
0.85
0.82
0.88
Support Vector Machine 0.78
0.75
0.81
0.78
0.85
Gradient Boosting
0.85
0.83
0.88
0.85
0.90
Neural Networks
0.88
0.86
0.89
0.87
0.91
Interpretation of Results
From the table, it is evident that Neural Networks achieved the highest performance across all evaluation metrics.
The accuracy of 88% indicates that the neural network model correctly predicted the clinical outcome in 88% of
the cases. The precision and recall scores of 0.86 and 0.89, respectively, show that the model was highly effective
in predicting both the positive and negative cases of improvement or worsening in patients undergoing laser
therapy for periodontal disease. The F1-Score of 0.87 further indicates the balance between precision and recall,
which is crucial for clinical settings where both false positives and false negatives can have significant implications.
Chart 1: Model Performance
The Gradient Boosting model also performed well, with
an accuracy of 85% and an AUC-ROC score of 0.90,
indicating that it had a strong ability to distinguish
between improved and worsened clinical outcomes.
The model demonstrated high precision (0.83) and
recall (0.88), making it a reliable option for predicting
treatment outcomes.
The Random Forest model provided a solid
performance with an accuracy of 82% and an AUC-ROC
score of 0.88. While its precision and recall were lower
than that of neural networks and gradient boosting, it
still offered robust predictive capabilities. The Random
Forest model is particularly useful due to its ability to
handle large datasets and provide insights into feature
importance, helping to identify key factors that
contribute to treatment success.
Support Vector Machine (SVM) showed decent
performance with an accuracy of 78% and an AUC-ROC
of 0.85. However, its precision and recall were slightly
lower than the top-performing models. SVM is
generally more suited for high-dimensional datasets,
and its performance may be improved further with
feature tuning or kernel adjustments.
The Logistic Regression model served as the baseline
and performed relatively well, with an accuracy of 75%.
However, its performance lagged behind the more
complex models, such as neural networks and gradient
boosting, in terms of precision, recall, and AUC-ROC.
0.75
0.82
0.78
0.85
0.88
0.72
0.8
0.75
0.83
0.86
0.78
0.85
0.81
0.88
0.89
0.75
0.82
0.78
0.85
0.87
0.82
0.88
0.85
0.9
0.91
L O G I S T I C
R E G R E S S I O N
R A N D O M F O R E S T
S U P P O R T V E C T O R
M A C H I N E
G R A D I E N T
B O O S T I N G
N E U R A L
N E T W O R K S
Accuracy
Precision
Recall
F1-Score
AUC-ROC
The American Journal of Medical Sciences and Pharmaceutical Research
33
https://www.theamericanjournals.com/index.php/tajmspr
The American Journal of Medical Sciences and Pharmaceutical Research
Chart 2: AUC_ROC curve for patient
The AUC-ROC curve (Area Under the Receiver
Operating Characteristic curve) is a graphical
representation of a model's ability to distinguish
between positive and negative outcomes. In our study,
the AUC-ROC was used to evaluate the performance of
machine
learning
models
in
predicting
the
effectiveness of laser therapy for periodontal disease
treatment. The curve plots the True Positive Rate
(sensitivity) against the False Positive Rate (1-
specificity) across different threshold values, providing
a comprehensive view of the model’s diagnostic
performance.
From the result table, the Neural Network achieved the
highest AUC-ROC score of 0.91, indicating its excellent
ability to differentiate between patients who are likely
to improve and those who may not benefit from laser
therapy. For example, a high AUC value signifies that
the model can accurately identify patients who will
respond positively to the treatment (true positives)
while minimizing incorrect predictions, such as
categorizing a non-responsive patient as responsive
(false positives).
In real-life clinical applications, the AUC-ROC helps
practitioners prioritize patients based on predicted
treatment success. For instance, a patient flagged by
the model with high confidence as likely to improve can
be prioritized for laser therapy, while alternative
treatments or additional diagnostic evaluations may be
recommended for patients with lower predicted
success rates. This ensures personalized treatment,
improves clinical outcomes, and optimizes healthcare
resources.
Key Insights and Clinical Relevance
The results indicate that machine learning models,
particularly neural networks and gradient boosting, can
effectively predict the outcomes of laser therapy in
treating periodontal diseases. The ability to predict
clinical improvements or deterioration post-treatment
can greatly assist clinicians in making informed
decisions about the treatment approach for individual
patients.
The feature importance analysis, which was conducted
using the Random Forest and Gradient Boosting
models, revealed that the most significant predictors of
treatment success were pre-treatment pocket depth,
laser therapy duration, and gingival inflammation level.
These variables were strongly associated with clinical
outcomes, suggesting that they play a pivotal role in
determining the effectiveness of laser therapy.
Moreover, the study highlighted that the Follow-Up
Period was also an important predictor of success, with
longer follow-up times correlating with better recovery
and more sustained improvements. This insight is
valuable for clinicians as it suggests that a longer post-
treatment monitoring period may contribute to better
outcomes.
Model Interpretability and Practical Application
While the neural network model provided the highest
accuracy, its complexity may limit interpretability.
However, the use of techniques like SHAP (Shapley
Additive Explanations) helped provide insights into how
different features contributed to individual predictions,
making the model more transparent and clinically
useful. The feature importance analysis from Random
The American Journal of Medical Sciences and Pharmaceutical Research
34
https://www.theamericanjournals.com/index.php/tajmspr
The American Journal of Medical Sciences and Pharmaceutical Research
Forest and Gradient Boosting also provided a clearer
understanding of the key factors that influence
treatment outcomes, offering actionable insights for
clinicians. The machine learning models evaluated in
this study successfully predicted the clinical outcomes
of laser therapy for periodontal diseases. Neural
networks, gradient boosting, and random forest
classifiers performed the best, with neural networks
emerging as the most effective model for predicting
treatment success. These results suggest that machine
learning techniques, when properly applied, can
significantly enhance the decision-making process in
periodontal care, leading to better patient outcomes
and more personalized treatment strategies.
The study also highlights the importance of factors such
as pre-treatment pocket depth, laser therapy duration,
and gingival inflammation in predicting clinical success.
These findings can guide clinicians in selecting the
optimal treatment parameters for individual patients,
ultimately improving the effectiveness of laser therapy
in periodontal disease treatment.
DISCUSSION
The findings of this study emphasize the potential of
machine learning in evaluating and predicting the
effectiveness of laser therapy for periodontal diseases.
By leveraging diverse machine learning models,
including Neural Networks, Gradient Boosting, and
Random Forest, we were able to achieve high levels of
predictive accuracy, with the Neural Network model
achieving the highest AUC-ROC of 0.91. These results
demonstrate that machine learning algorithms can
effectively process clinical and demographic data to
identify patterns and predict treatment outcomes with
precision.
The application of machine learning in this context
highlights its ability to identify key clinical variables,
such
as
periodontal
depth,
gingival
inflammation, and patient age, which significantly
influence the success of laser therapy. This not only
validates the effectiveness of laser therapy but also
supports the use of predictive modeling to enhance
treatment personalization. For instance, clinicians can
use these predictive insights to determine the
likelihood of success for individual patients, enabling
more targeted and efficient therapeutic interventions.
Moreover, the study confirms that laser therapy
remains an effective non-invasive treatment option for
managing periodontal diseases, particularly when
coupled with data-driven insights. However, it is
essential to note that model performance can vary
based on data quality, feature selection, and the
representativeness of the dataset. Therefore, while
machine learning provides valuable predictive tools,
clinical judgment and patient-specific factors should
remain central to treatment planning.
Future studies should focus on larger and more diverse
datasets to further validate the generalizability of these
models. Additionally, incorporating other clinical
variables, such as genetic markers or microbiological
data, could enhance predictive accuracy and provide
deeper insights into treatment mechanisms.
CONCLUSION
This study successfully demonstrates the application of
machine learning models to predict the effectiveness of
laser therapy in treating periodontal diseases. By
evaluating various algorithms, we identified Neural
Networks and Gradient Boosting as the most effective
models, achieving AUC-ROC scores of 0.91 and 0.90,
respectively. These results underline the potential of
machine learning to revolutionize clinical decision-
making by providing accurate, data-driven predictions.
The integration of machine learning into periodontal
care can improve treatment outcomes by enabling
clinicians to personalize therapy plans based on
individual patient characteristics. Laser therapy,
supported by predictive modeling, emerges as a
promising approach to managing periodontal diseases
with improved precision and efficiency.
In conclusion, this study bridges the gap between
advanced technologies and clinical practice, offering a
framework for incorporating machine learning into
evidence-based dentistry. While our findings are
promising, further research with larger datasets and
additional clinical variables is needed to refine these
predictive models and expand their applicability. The
combined use of machine learning and laser therapy
has the potential to set a new standard for patient-
centered periodontal care, driving better outcomes for
patients and advancing the field of dental medicine.
Acknowledgement:
All the Authors contributed equally
REFERENCE
Md Habibur Rahman, Ashim Chandra Das, Md Shujan
Shak, Md Kafil Uddin, Md Imdadul Alam, Nafis Anjum,
Md Nad Vi Al Bony, & Murshida Alam. (2024).
TRANSFORMING CUSTOMER RETENTION IN FINTECH
INDUSTRY THROUGH PREDICTIVE ANALYTICS AND
MACHINE LEARNING. The American Journal of
Engineering
and
Technology,
6(10), 150
–
163.
https://doi.org/10.37547/tajet/Volume06Issue10-17
Han, S. S., Kim, M. S., Lim, W., Park, G. H., Park, I., &
Chang, S. E. (2020). Classification of the clinical images
for benign and malignant cutaneous tumors using a
deep learning algorithm. Journal of Investigative
Dermatology,
140(7),
1572-1579.
https://doi.org/10.1016/j.jid.2019.12.008
The American Journal of Medical Sciences and Pharmaceutical Research
35
https://www.theamericanjournals.com/index.php/tajmspr
The American Journal of Medical Sciences and Pharmaceutical Research
Hekler, A., Utikal, J. S., Enk, A. H., Berking, C., Klode, J.,
Hauschild, A., ... & von Kalle, C. (2019). Superior skin
cancer classification by the combination of human and
artificial intelligence. European Journal of Cancer, 120,
114-121. https://doi.org/10.1016/j.ejca.2019.07.013
Nasr-Esfahani, E., Samavi, S., Karimi, N., Soroushmehr,
S. M., Jafari, M. H., & Ward, K. (2018). Melanoma
detection by analysis of clinical images using
convolutional neural networks. Artificial Intelligence in
Medicine,
87,
54-63.
https://doi.org/10.1016/j.artmed.2018.03.008
Tauhedur Rahman, Md Kafil Uddin, Biswanath
Bhattacharjee, Md Siam Taluckder, Sanjida Nowshin
Mou, Pinky Akter, Md Shakhaowat Hossain, Md Rashel
Miah, & Md Mohibur Rahman. (2024). BLOCKCHAIN
APPLICATIONS IN BUSINESS OPERATIONS AND SUPPLY
CHAIN MANAGEMENT BY MACHINE LEARNING.
International Journal of Computer Science &
Information
System,
9(11),
17
–
30.
https://doi.org/10.55640/ijcsis/Volume09Issue11-03
Md Jamil Ahmmed, Md Mohibur Rahman, Ashim
Chandra Das, Pritom Das, Tamanna Pervin, Sadia Afrin,
Sanjida Akter Tisha, Md Mehedi Hassan, & Nabila
Rahman. (2024). COMPARATIVE ANALYSIS OF
MACHINE LEARNING ALGORITHMS FOR BANKING
FRAUD DETECTION: A STUDY ON PERFORMANCE,
PRECISION,
AND
REAL-TIME
APPLICATION.
International Journal of Computer Science &
Information
System,
9(11),
31
–
44.
https://doi.org/10.55640/ijcsis/Volume09Issue11-04
Nafis Anjum, Md Nad Vi Al Bony, Murshida Alam,
Mehedi Hasan, Salma Akter, Zannatun Ferdus, Md
Sayem Ul Haque, Radha Das, & Sadia Sultana. (2024).
COMPARATIVE ANALYSIS OF SENTIMENT ANALYSIS
MODELS ON BANKING INVESTMENT IMPACT BY
MACHINE LEARNING ALGORITHM. International
Journal of Computer Science & Information System,
9(11),
5
–
16.
https://doi.org/10.55640/ijcsis/Volume09Issue11-02
Shahbazi, Z., & Byun, Y. C. (2021). Integration of
blockchain, IoT and machine learning for multistage
quality control and enhancing security in smart
manufacturing. Sensors, 21(4), 1467.
Das, A. C., Mozumder, M. S. A., Hasan, M. A., Bhuiyan,
M., Islam, M. R., Hossain, M. N., ... & Alam,
Al Mamun, A., Hossain, M. S., Rishad, S. S. I., Rahman,
M. M., Shakil, F., Choudhury, M. Z. M. E., ... & Sultana,
S. (2024). MACHINE LEARNING FOR STOCK MARKET
SECURITY MEASUREMENT: A COMPARATIVE ANALYSIS
OF SUPERVISED, UNSUPERVISED, AND DEEP LEARNING
MODELS. The American Journal of Engineering and
Technology, 6(11), 63-76.
Haque, M. S., Amin, M. S., Ahmad, S., Sayed, M. A.,
Raihan, A., & Hossain, M. A. (2023, September).
Predicting Kidney Failure using an Ensemble Machine
Learning Model: A Comparative Study. In 2023 10th
International Conference on Electrical Engineering,
Computer Science and Informatics (EECSI) (pp. 31-37).
IEEE.
Modak, C., Shahriyar, M. A., Taluckder, M. S., Haque, M.
S., & Sayed, M. A. (2023, August). A Study of Lung
Cancer Prediction Using Machine Learning Algorithms.
In 2023 3rd International Conference on Electronic and
Electrical Engineering and Intelligent System (ICE3IS)
(pp. 213-217). IEEE.
Modak, C., Shahriyar, M. A., Taluckder, M. S., Haque, M.
S., & Sayed, M. A. (2023, August). A Study of Lung
Cancer Prediction Using Machine Learning Algorithms.
In 2023 3rd International Conference on Electronic and
Electrical Engineering and Intelligent System (ICE3IS)
(pp. 213-217). IEEE.
Nguyen, T. N., Khan, M. M., Hossain, M. Z., Sharif, K. S.,
Das, R., & Haque, M. S. (2024). Product Demand
Forecasting For Inventory Management with Freight
Transportation Services Index Using Advanced Neural
Networks Algorithm. American Journal of Computing
and Engineering, 7(4), 50-58.
Miah, J., Khan, R. H., Ahmed, S., & Mahmud, M. I. (2023,
June). A comparative study of detecting covid 19 by
using chest X-ray images
–
A deep learning approach. In
2023 IEEE World AI IoT Congress (AIIoT) (pp. 0311-
0316). IEEE.
Khan, R. H., Miah, J., Nipun, S. A. A., & Islam, M. (2023,
March). A Comparative Study of Machine Learning
classifiers to analyze the Precision of Myocardial
Infarction prediction. In 2023 IEEE 13th Annual
Computing and Communication Workshop and
Conference (CCWC) (pp. 0949-0954). IEEE.
Kayyum, S., Miah, J., Shadaab, A., Islam, M. M., Islam,
M., Nipun, S. A. A., ... & Al Faisal, F. (2020, January).
Data analysis on myocardial infarction with the help of
machine learning algorithms considering distinctive or
non-distinctive features. In 2020 International
Conference on Computer Communication and
Informatics (ICCCI) (pp. 1-7). IEEE.
Islam, M. M., Nipun, S. A. A., Islam, M., Rahat, M. A. R.,
Miah, J., Kayyum, S., ... & Al Faisal, F. (2020). An
empirical study to predict myocardial infarction using k-
means and hierarchical clustering. In Machine Learning,
Image Processing, Network Security and Data Sciences:
Second International Conference, MIND 2020, Silchar,
India, July 30-31, 2020, Proceedings, Part II 2 (pp. 120-
130). Springer Singapore.
Miah, J., Ca, D. M., Sayed, M. A., Lipu, E. R., Mahmud,
The American Journal of Medical Sciences and Pharmaceutical Research
36
https://www.theamericanjournals.com/index.php/tajmspr
The American Journal of Medical Sciences and Pharmaceutical Research
F., & Arafat, S. Y. (2023, November). Improving
Cardiovascular
Disease
Prediction
Through
Comparative Analysis of Machine Learning Models: A
Case Study on Myocardial Infarction. In 2023 15th
International Conference on Innovations in Information
Technology (IIT) (pp. 49-54). IEEE.
Khan, R. H., Miah, J., Rahat, M. A. R., Ahmed, A. H.,
Shahriyar, M. A., & Lipu, E. R. (2023, September). A
Comparative Analysis of Machine Learning Approaches
for Chronic Kidney Disease Detection. In 2023 8th
International Conference on Electrical, Electronics and
Information Engineering (ICEEIE) (pp. 1-6). IEEE.
Miah, J., Cao, D. M., Sayed, M. A., Taluckder, M. S.,
Haque, M. S., & Mahmud, F. (2023). Advancing Brain
Tumor Detection: A Thorough Investigation of CNNs,
Clustering, and SoftMax Classification in the Analysis of
MRI Images. arXiv preprint arXiv:2310.17720.
Rahman, M. M., Islam, A. M., Miah, J., Ahmad, S., &
Mamun, M. (2023, June). sleepWell: Stress Level
Prediction Through Sleep Data. Are You Stressed?. In
2023 IEEE World AI IoT Congress (AIIoT) (pp. 0229-
0235). IEEE.
Rahman, M. M., Islam, A. M., Miah, J., Ahmad, S., &
Hasan, M. M. (2023, June). Empirical Analysis with
Component Decomposition Methods for Cervical
Cancer Risk Assessment. In 2023 IEEE World AI IoT
Congress (AIIoT) (pp. 0513-0519). IEEE.
Khan, R. H., Miah, J., Nipun, S. A. A., Islam, M., Amin, M.
S., & Taluckder, M. S. (2023, September). Enhancing
Lung Cancer Diagnosis with Machine Learning Methods
and Systematic Review Synthesis. In 2023 8th
International Conference on Electrical, Electronics and
Information Engineering (ICEEIE) (pp. 1-5). IEEE.
Miah, J. (2024). HOW FAMILY DNA CAN CAUSE LUNG
CANCER USING MACHINE LEARNING. International
Journal of Medical Science and Public Health Research,
5(12), 8-14.
Miah, J., Khan, R. H., Linkon, A. A., Bhuiyan, M. S., Jewel,
R. M., Ayon, E. H., ... & Tanvir Islam, M. (2024).
Developing a Deep Learning Methodology to Anticipate
the Onset of Diabetic Retinopathy at an Early Stage. In
Innovative and Intelligent Digital Technologies;
Towards an Increased Efficiency: Volume 1 (pp. 77-91).
Cham: Springer Nature Switzerland.
M. I. (2024). MACHINE LEARNING APPROACHES FOR
DEMAND FORECASTING: THE IMPACT OF CUSTOMER
SATISFACTION ON PREDICTION ACCURACY. The
American Journal of Engineering and Technology, 6(10),
42-53.
Ontor, M. R. H., Iqbal, A., Ahmed, E., & Rahman, A.
LEVERAGING DIGITAL TRANSFORMATION AND SOCIAL
MEDIA ANALYTICS FOR OPTIMIZING US FASHION
BRANDS'PERFORMANCE: A MACHINE LEARNING
APPROACH. SYSTEM (eISSN: 2536-7919 pISSN: 2536-
7900), 9(11), 45-56.
Hossain, M. N., Hossain, S., Nath, A., Nath, P. C., Ayub,
M. I., Hassan, M. M., ... & Rasel, M. (2024). ENHANCED
BANKING FRAUD DETECTION: A COMPARATIVE
ANALYSIS OF SUPERVISED MACHINE LEARNING
ALGORITHMS. American Research Index Library, 23-35.
Shak, M. S., Mozumder, M. S. A., Hasan, M. A., Das, A.
C., Miah, M. R., Akter, S., & Hossain, M. N. (2024).
OPTIMIZING RETAIL DEMAND FORECASTING: A
PERFORMANCE EVALUATION OF MACHINE LEARNING
MODELS INCLUDING LSTM AND GRADIENT BOOSTING.
The American Journal of Engineering and Technology,
6(09), 67-80.
Das, A. C., Mozumder, M. S. A., Hasan, M. A., Bhuiyan,
M., Islam, M. R., Hossain, M. N., ... & Alam, M. I. (2024).
MACHINE LEARNING APPROACHES FOR DEMAND
FORECASTING:
THE
IMPACT
OF
CUSTOMER
SATISFACTION ON PREDICTION ACCURACY. The
American Journal of Engineering and Technology, 6(10),
42-53.
Hossain, M. N., Anjum, N., Alam, M., Rahman, M. H.,
Taluckder, M. S., Al Bony, M. N. V., ... & Jui, A. H. (2024).
PERFORMANCE OF MACHINE LEARNING ALGORITHMS
FOR LUNG CANCER PREDICTION: A COMPARATIVE
STUDY. International Journal of Medical Science and
Public Health Research, 5(11), 41-55.
Rahman, A., Iqbal, A., Ahmed, E., & Ontor, M. R. H.
(2024). PRIVACY-PRESERVING MACHINE LEARNING:
TECHNIQUES, CHALLENGES, AND FUTURE DIRECTIONS
IN SAFEGUARDING PERSONAL DATA MANAGEMENT.
International journal of business and management
sciences, 4(12), 18-32.
Md Jamil Ahmmed, Md Mohibur Rahman, Ashim
Chandra Das, Pritom Das, Tamanna Pervin, Sadia Afrin,
Sanjida Akter Tisha, Md Mehedi Hassan, & Nabila
Rahman. (2024). COMPARATIVE ANALYSIS OF
MACHINE LEARNING ALGORITHMS FOR BANKING
FRAUD DETECTION: A STUDY ON PERFORMANCE,
PRECISION,
AND
REAL-TIME
APPLICATION.
International Journal of Computer Science &
Information
System,
9(11),
31
–
44.
https://doi.org/10.55640/ijcsis/Volume09Issue11-04
Arif, M., Ahmed, M. P., Al Mamun, A., Uddin, M. K.,
Mahmud, F., Rahman, T., ... & Helal, M. (2024).
DYNAMIC PRICING IN FINANCIAL TECHNOLOGY:
EVALUATING MACHINE LEARNING SOLUTIONS FOR
MARKET ADAPTABILITY. International Interdisciplinary
Business Economics Advancement Journal, 5(10), 13-
27.
Uddin, M. K., Akter, S., Das, P., Anjum, N., Akter, S.,
The American Journal of Medical Sciences and Pharmaceutical Research
37
https://www.theamericanjournals.com/index.php/tajmspr
The American Journal of Medical Sciences and Pharmaceutical Research
Alam, M., ... & Pervin, T. (2024). MACHINE LEARNING-
BASED EARLY DETECTION OF KIDNEY DISEASE: A
COMPARATIVE STUDY OF PREDICTION MODELS AND
PERFORMANCE EVALUATION. International Journal of
Medical Science and Public Health Research, 5(12), 58-
75.
Das, A. C., Rishad, S. S. I., Akter, P., Tisha, S. A., Afrin, S.,
Shakil, F., ... & Rahman, M. M. (2024). ENHANCING
BLOCKCHAIN SECURITY WITH MACHINE LEARNING: A
COMPREHENSIVE STUDY OF ALGORITHMS AND
APPLICATIONS. The American Journal of Engineering
and Technology, 6(12), 150-162.
Shak, M. S., Uddin, A., Rahman, M. H., Anjum, N., Al
Bony, M. N. V., Alam, M., ... & Pervin, T. (2024).
INNOVATIVE MACHINE LEARNING APPROACHES TO
FOSTER FINANCIAL INCLUSION IN MICROFINANCE.
International Interdisciplinary Business Economics
Advancement Journal, 5(11), 6-20.
Bhattacharjee, B., Mou, S. N., Hossain, M. S., Rahman,
M. K., Hassan, M. M., Rahman, N., ... & Haque, M. S. U.
(2024). MACHINE LEARNING FOR COST ESTIMATION
AND FORECASTING IN BANKING: A COMPARATIVE
ANALYSIS OF ALGORITHMS. Frontline Marketing,
Management and Economics Journal, 4(12), 66-83.
Rahman, A., Iqbal, A., Ahmed, E., & Ontor, M. R. H.
(2024). PRIVACY-PRESERVING MACHINE LEARNING:
TECHNIQUES, CHALLENGES, AND FUTURE DIRECTIONS
IN SAFEGUARDING PERSONAL DATA MANAGEMENT.
Frontline Marketing, Management and Economics
Journal, 4(12), 84-106.
Rahman, M. M., Akhi, S. S., Hossain, S., Ayub, M. I.,
Siddique, M. T., Nath, A., ... & Hassan, M. M. (2024).
EVALUATING MACHINE LEARNING MODELS FOR
OPTIMAL CUSTOMER SEGMENTATION IN BANKING: A
COMPARATIVE STUDY. The American Journal of
Engineering and Technology, 6(12), 68-83.
Das, P., Pervin, T., Bhattacharjee, B., Karim, M. R.,
Sultana, N., Khan, M. S., ... & Kamruzzaman, F. N. U.
(2024). OPTIMIZING REAL-TIME DYNAMIC PRICING
STRATEGIES IN RETAIL AND E-COMMERCE USING
MACHINE LEARNING MODELS. The American Journal of
Engineering and Technology, 6(12), 163-177.
