Fast and High Precision Human Fall Prevention and Intimation Using

Haptic Technology

R. Ravichandran, B. Latha, B. Vinothkumar, K. Arunkarthik, N G. Gopisankar and A. Yuvarajan

Department of Electronics and Communication Engineering, K.S.R. College of Engineering, Tiruchengode, Namakkal,

Tamil Nadu, India

Keywords: Fall Detection, Haptic Technology, YOLOv8, Human Fall Prevention, Response Time, Accuracy, False

Positive Rate, Wearable Sensors.

Abstract: The aim of the study is to develop a fast and high-precision human fall prevention and intimation system

using haptic technology, comparing its performance with the existing YOLOv8-based fall detection method.

Materials and Methods: There are two groups in this study. Group 1 refers to a novel YOLOv8-based fall

detection approach with 12-samples and Group 2 refers to a haptic- based fall detection method with 12-

samples. The threshold is 0.05% with a 95% confidence range and the G Power value is 80%. Result: The

performance of the proposed haptic-based fall detection system is noticeably superior to that of the YOLOv8-

based technique. Whereas the YOLOv8-based approach attains an accuracy between 91.50% and 95.30%,

the haptic-based method's accuracy falls between 95.80% and 98.50%. At a significance level of p < 0.05,

the haptic-based solution performs at its best. Conclusion: In this work, it is observed that the haptic- based

fall detection system has significantly better accuracy and response time compared to the YOLOv8- based

method.

1 INTRODUCTION

M. A. Kuhail, Fast and High Precision Human Fall

Prevention and Intimation using Haptic Technology,

a wearable system, detects fall-like symptoms and

generates haptic feedback alerts to prevent falls

using motion sensors and predictive algorithms. It

monitors critical health metrics for improved safety

and well-being and, in the event of a fall, initiates

emergency warnings through IoT connectivity.

Based on an enhanced YOLOv8 framework, the

FDT- YOLO algorithm tackles fall detection issues

by substituting the Faster Net module for the C2f

module, which improves feature reuse and reduces

complexity. T. Yang., et al, 2024 To further develop

precision and strength, it incorporates a trio

consideration strategy to lessen foundation

impedance and a deformable convolution module to

deal with complex poses. The system achieves a

higher Aide (96.2 % at IoU 0.5) while diminishing

limits to 9.9 million, further creating both ID

exactness and efficiency. Falls address a colossal

prosperity risk for more prepared adults, driving the

necessity for state-of-the-art fall area and countering

structures. Customary wearable methodologies

frequently depend on factual techniques with high

deception rates. H. Jing et al., 2025; Singh, 2023 This

paper overviews progressing designs in fall

acknowledgment and contravention using computer

based intelligence (ML), looking at datasets, age get-

togethers, sensors, and computations, while

highlighting stream movies and future headings to

coordinate researchers in additional fostering these

systems. L. Liu, et al., 2025 The structure can be used

in homes or aided living workplaces to prevent falls

among more prepared adults, giving nonstop alerts

and prosperity checking to ensure their security and

well-being. C.-W. Kang, et al., 2025 Hospitals and

reclamation centers can convey the structure for

patients with flexibility issues or postoperative

recovery, offering proactive fall balance and speedy

emergency response in case of a fall.

2 RELATED WORKS

With more than 300 conveyances in IEEE Xplore,

151 papers in Google Scholar, and 123 in

632

Ravichandran, R., Latha, B., Vinothkumar, B., Arunkarthik, K., Gopisankar, N. G. and Yuvarajan, A.

Fast and High Precision Human Fall Prevention and Intimation Using Haptic Technology.

DOI: 10.5220/0013903200004919

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 3, pages

632-638

ISBN: 978-989-758-777-1

academia.edu, fall neutralization and recognizable

proof has seen a sharp extension in interest all

through late years. To address the colossal

impediments looked by the old and patients with

compactness weaknesses, J. Stuby, et al., 2025; AS

Kalyana Kumar, et.al, 2025 this assessment presents a

unique wearable contraption for fall countering and

acknowledgment that organizes development sensors,

Web of Things developments, and progressing

prosperity checking. E. Cubo et al., 2025 The

prescribed course of action uses insightful

computations to review development models and

combines whirligigs and accelerometers to follow

client developments. C. Chen, et al., 2025 The

contraption makes haptic analysis alerts to ask

clients to settle themselves to thwart falls, expecting

it recognizes possible fall side effects. D. Gnutt et

al., 2019; Banu, in case of a genuine fall, an

emergency cautioning structure is set off, sending the

client's district and achievement data to parental

figures through IoT stages. Prospering evaluations,

for instance, are constantly moved to the Think Speak

IoT stage, giving far off enrollment to gatekeepers

and family members. T. Ishigaki et al., 2025 The design

has been normal for lightweight, wearable use and

accomplishes an affirmation accuracy of 85-95 %,

with savvy limits diminishing confusions essentially.

J. Gudin, et al., 2025 Not a tiny smidgen like standard

designs, which depend upon camera-based

distinctive verification or edge-based wearables, this

gadget gives proactive assumption through its two-

stage structure. K. Pacheco-Barrios, et al., 2024 The

utilization of IoT union guarantees that crisis

reactions are quick and cautious, limiting defers that

could incite crazy results. Meenakshi Sharma, et.al,

2023; A. Musa, 2024 To redesign execution, the

sensible calculations are further developed utilizing

progressed signal dealing with techniques,

guaranteeing constant development with low

dormancy. J. Chadokiya et al., 2024 This try watches

out for an enormous jump forward in wearable turn

of events, cementing fall repudiation, ID, and

flourishing seeing into a solitary, helpful contraption

that watches out for both the security and flourishing

of clients. X. Hu, et al., 2025 With its versatility, the

framework can be applied in old homes, clinical

offices, and recovery focuses, making it a principal

device for working on individual satisfaction in

weak people groups.

From the past disclosures, it is contemplated that

standard fall acknowledgment systems have

limitations in accuracy and responsiveness, every

now and again achieving high deceptive issues.

Enhancing fall location and counteraction is a vital

component while planning wearable wellbeing

frameworks. The point of this study is to improve

fall counteraction and location precision utilizing a

movement sensor-based prescient framework with

haptic criticism, in correlation with customary limit-

based frameworks.

3 MATERIALS AND METHODS

The review was directed in the KSRIET Association

Lab utilizing movement information and biometric

wellbeing boundaries acquired from wearable

sensors coordinated with the IoT stage Think Speak.

A sum of 500 movement and wellbeing information

tests were handled involving very much planned

structures and modified involving C language for

framework improvement. This task underscores high

precision in fall avoidance and recognition by using

movement sensors, prescient calculations, and haptic

criticism to guarantee client wellbeing and

convenient mediations. Singh, 2023; N. T. Newaz and

E. Hanada, 2025 The framework joins movement

examination and wellbeing checking to upgrade fall

recognition, essentially decreasing misleading

problems. Information gathered from

accelerometers, spinners, and pulse sensors were

handled progressively utilizing C programming,

guaranteeing productive and quick execution of the

fall identification and warning framework. With a

reaction precision of 85-95 % and a streamlined

location edge, the framework gives solid fall

counteraction and recognition, further developing

crisis reaction times and generally speaking client

security. The reconciliation of C language

programming with IoT guarantees powerful and

ongoing framework execution, making it a useful

answer for improving security and prosperity,

particularly for old people and patients.

Group 1 refers to the fall detection method using

YOLOv8. The framework streamlines for

boundaries like objective and handling speed while

processing image or video input for continuous fall

discovery. To ensure precise placement under various

conditions, the calculation is prepared using fall-

related datasets and evaluated based on Accuracy and

Review. M. Szántó et al., 2025 The fall

identification and crisis response framework, which

uses the IoT mix and wellness checking highlights to

process information.

Group 2 refers to the fall identification and crisis

reaction framework, which processes information

utilizing IoT mix and wellbeing checking highlights.

The framework examines essential signs, for

Fast and High Precision Human Fall Prevention and Intimation Using Haptic Technology

633

example, pulse and movement information, sending

continuous cautions in case of a fall. It works

utilizing an associated stage like Think Speak and is

enhanced for crisis warnings, guaranteeing quick

reaction times. The framework aspects are intended

for consistent incorporation into wearable devices.

The framework is planned involving the

accompanying condition for fall recognition and

anticipation is defined by Equation (1)

𝑆  1/𝑁 ∑𝑚𝑖 (1)

Here, S is the framework's reaction precision, N

addresses the quantity of movement tests

investigated, and mi alludes to the singular

movement test information from accelerometers and

gyrators. The framework works utilizing a scope of

movement examples and wellbeing checking

information, such as heart rate, to foresee falls and

produce constant cautions.

The testing arrangement and reenactment of the

fall location and avoidance framework are designed as

follows: The framework is executed utilizing an

eighth Gen Intel i7 centre processor, 8 GB Slam, and

modified in C language. The framework is intended

to handle motion sensor information (accelerometers

and Gyroscopes) continuously, involving predictive

algorithms and haptic feedback for fall location. The

information input is handled through the Think Speak

IoT stage for remote checking and ready age. The

framework's exactness is approved by applying

ongoing movement information, breaking down the

outcomes, and upgrading the prescient calculations

to further develop fall location and decrease phony

problems. The fall discovery framework is then

recreated and broken down for continuous execution

and responsiveness.

Figure 1: Flowchart of detection and health monitoring

system.

Figure 1: This flowchart demonstrates how fall

detection and health monitoring systems operate by

employing motion sensors to continuously track

human motions. The system alerts the user by

vibrating if it detects a possible fall. An vibrating if

it detects a possible fall. An emergency notification

with the user's location is issued if the fall is verified

and no response is received from them within a

predetermined period of time. The technology

simultaneously tracks vital signs including heart rate

and notifies users if any irregularities are found. It

keeps an eye on activity and health metrics if there

are no problems.

4 STATISTICAL ANALYSIS

SPSS version 26.0 is used for statistical analysis of

data collected from parameters such as accuracy (%)

and response time (s). O. Korostynska and A. Mason,

2021 The independent sample t-test and group

statistics are calculated using SPSS software. The

sensor placement, algorithm time and feedback

mechanism are considered independent variables,

while accuracy (%) and response time (s) are the

dependent variables.

5 RESULTS

The results of the novel haptic-based fall prediction

model's results against YOLOv8 are done. The two

models' accuracy, response time, and false positive

rate (FPR) are examined. While the suggested haptic

prediction model has an enhanced accuracy range of

93.80% to 98.50%, the accuracy values for YOLOv8

range from 91.50% to 95.30%. The haptic-based

system showed lower FPR values (0.22 to 0.33) than

YOLOv8 (0.39 to 0.50) and better response times

ranging from 0.60s to 0.85s, whereas YOLOv8

exhibited response times ranging from 1.00s to

1.20s. The associated changes in FPR and response

time were also measured.

The accuracy ranges for YOLOv8 and haptic

prediction are compared to a maximum of 95.30%

and 98.50%, respectively. For haptic prediction, the

minimal accuracy range is 93.80%, while for

YOLOv8, it is 91.50%. Table 1 presents the results

of the independent sample t-test comparison of

haptic prediction and YOLOv8 accuracy. Table 2

displays the accuracy mean, standard deviation, and

significance levels for both models. Table 3

(independent sample t-test) indicates that the

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

COMMUNICATION, AND COMPUTING TECHNOLOGIES

634

accuracy improvement with the haptic prediction

model was statistically significant (p < 0.05).

Table 1: The accuracy ranges from 91.50 % to

95.30 % for the yolov8 model and 95.80 % to 98.50

% for the haptic prediction model, demonstrating

significant improvement in accuracy with the

proposed method. The false positive rate(fpr)

decreases from 0.50 to 0.39 in yolov8 and 0.33 to

0.22 in haptic prediction. Additionally, the response

time is reduced from 1.20s to 1.00s in yolov8 and

0.85s to 0.60s in haptic prediction, ensuring faster

fall detection and prevention.

Table 2. T-test in yolov8 n is 12 and mean value is

93.8917, with s std. Deviation of 1.84068 and a std.

Error mean of 0.53136.for haptic prediction,the

mean value is 97.3000, with a std. Deviation of

0.95537 and a std.error mean of 0.27579.

The mean, standard deviation, and significant

difference of accuracy between the YOLOv8 and

Haptic prediction are classified in Table 3, which

demonstrates a significant difference between the

two groups (p < 0.05, independent sample t-test).

Figures and Tables

Table 1: Comparison of Performance Metrics between YOLOv8 and Haptic Prediction across 12 Iterations

Iteration Accuracy FPR Response Time

YOLOv

Haptic

Prediction

YOLOv

Haptic

Prediction

YOLOv

Haptic

Prediction

1 92.80 96.20 0.46 0.30 1.10 0.80

94.50 97.00 0.42 0.28 1.00 0.75

3 92.80 93.80 0.49 0.33 1.20 0.85

95.00 98.00 0.44 0.27 1.00 0.70

5 93.50 96.50 0.47 0.31 1.10 0.78

93.00 98.20 0.40 0.25 1.00 0.65

7 92.30 94.50 0.48 0.29 1.10 0.72

8 94.80 98.00 0.43 0.26 1.00 0.68

9 95.30 98.50 0.39 0.22 1.00 0.60

10 91.50 96.00 0.50 0.32 1.20 0.80

11 95.30 98.10 0.41 0.24 1.10 0.64

12 93.00 97.80 0.47 0.28 1.20 0.70

Table 2: Summary Statistics for YOLOv8 and Haptic Prediction Models

Types of Models N Mean Std. Deviation Std.Error Mean

YOLOv8 1

93.8917 1.84068 0.53136

Haptic prediction 1

97.3000 0.95537 0.27579

Table 3. Independent sample test. T-test comparison with YOLOv8 and Haptic prediction (p<0.05).

Inde

endent Sam

le Test

Levene’s Test for

Equality of

Variances

t-test For Equality of Means

95%

Confidence interval

of Difference

F Sig. t df sig. (2-

tailed

)

Mean

Difference

Std error

Differenc

Lower Upper

Equal

variances

assumed

6.3

0.0

5.6

.000 -

3.4083

0.5986

4.649

2.166

Equal

variances

not

assumed

5.6

.000 -

3.4083

0.5986

4.674

2.142

Fast and High Precision Human Fall Prevention and Intimation Using Haptic Technology

635

The system architecture of the fall prevention

and intimation system using haptic technology is

shown in Figure 2. Similarly, the implementation and

data visualization of the fall prevention and

detection system is shown in Figure 3. Haptic

Prediction delivers the accuracy of 98.50% when

compared to YOLOv8’s 95.30% is shown in Figure

4 and Figure 5. False positive rate between YOLOv8

and Haptic Prediction models over multiple

iterations is shown in Figure 6. in which Haptic

Prediction demonstrates lower fluctuating false

positive rates of 0.33. Haptic Prediction

demonstrates higher response time of 0.60 Sec when

compared to YOLOv8 is shown in Figure 7.

Figure 2: The system architecture of the fall prevention

and intimation system using haptic technology. It

integrates an Arduino Nano with sensors for fall detection,

pulse, and temperature monitoring. Alerts are sent via a

buzzer, LCD, vibration motor, and GSM/GPS modules.

Figure 3: The implementation and data visualization of the

fall prevention and detection system. It includes hardware

setup with LCD displays (a) and real-time monitoring

graphs for body temperature, heart rate, and acceleration

in X and Y axes (b – e) using ThinkSpeak. These

visualizations help in analyzing fall patterns and system

responses.

Figure 4: This Bar Graph represents the accuracy between

YOLOv8 and Haptic Prediction with an accuracy of

98.50% when compared to YOLOv8’s 95.30%.

Figure 5: The graphs compare the accuracy of YOLOv8

and Haptic Prediction models over multiple iterations.

Haptic Prediction demonstrates higher accuracy compared

to the Haptic Prediction models.

Figure 6: The graph compares False positive rate between

YOLOv8 and Haptic Prediction models over multiple

iterations. Haptic Prediction demonstrates lower

fluctuating false positive rates of 0.33.

Figure 7: This graph compares the response time of

YOLOv8 and Haptic Prediction models across multiple

iterations. Haptic Prediction demonstrates higher response

time of 0.60 Sec.

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

COMMUNICATION, AND COMPUTING TECHNOLOGIES

636

6 DISCUSSIONS

The proposed haptic-based fall forecast framework

exhibits fundamentally preferable precision and

responsiveness over the current YOLOv8-based fall

discovery technique. Utilizing an autonomous

example T-test, the haptic framework shows further

developed fall expectation productivity by breaking

down movement sensor information progressively.

The framework is streamlined at a particular

recurrence reach to upgrade recognition accuracy

and diminish reaction time. The outcomes acquired

in this exploration show higher precision and

quicker cautions contrasted with previous studies.

The total gain obtained for haptic-based fall

identification technique is 98.5 % and for YOLOv8-

based fall identification technique is 95.3 %. A

precision improvement of around 4.5 % is achieved.

C.Rajan, et.al, 2023; W. Kemmler, 2020 A novel

design of a haptic-based fall prevention system is

created to improve fall prediction accuracy and

guarantee continuous client wellbeing. The

consequences of the proposed technique show

further developed fall location precision of 98.5 %

and diminishing misleading problems and improving

reaction time. M. Tavakoli, 2022 The framework

incorporates accelerometer and whirligig sensors

with cutting edge prescient calculations to recognize

fall-like developments. The proposed haptic input

instrument will offer additional opportunities for the

advancement of superior execution continuous fall

avoidance frameworks. Pradeep, R., 2019; A. Bohr

and K. Memarzadeh, 2020 For further developed

responsiveness, a minimal wearable gadget with

movement sensors and haptic actuators is conceived.

The proposed framework uses a mix of sensor

combination and AI to precisely order

developments. The framework presents a continuous

checking arrangement utilizing a mix of prescient

examination and haptic cautions for fall

counteraction applications. The movement sensors

constantly track client developments, while the

prescient calculation distinguishes fall dangers and

triggers cautions. Tareq Z. Ahram and Christianne S.

Falcão, 2022 The fall anticipation framework comes

in different wearable structures, including

wristbands, belts, and shrewd footwear. Wearable

solutions with integrated haptic criticism guarantee

better adherence and proactive fall anticipation.

The limitations of this plan remember the reliance

for sensor exactness and possible misleading up-

sides in identifying fall-like developments utilizing

the proposed haptic-based fall counteraction

framework. Because of the intricacy of constant

movement examination, execution time might be

longer while handling huge datasets. The

framework's wearable nature guarantees

convenience, smaller size, and reasonableness,

making it appropriate for older considerations,

medical clinic observing, and modern wellbeing

applications. In future examinations might

investigate progressed sensor combination strategies

and artificial intelligence driven prescient models to

upgrade fall identification precision and lessen

reaction time.

7 CONCLUSIONS

The fall detection system including the current

YOLOv8 model and the proposed haptic-based

forecast technique was planned and broke down.

The exactness of the proposed haptic-based

framework is fundamentally better compared to the

YOLOv8 model in foreseeing falls utilizing

continuous sensor information. The YOLOv8 model

accomplished a precision going from 91.5 % to 95.3

%, while the haptic-based forecast technique showed

improved exactness from 93.8 % to 98.5 %. The

standard deviation for the YOLOv8 technique is

1.84068, while the standard deviation for the

proposed haptic-based expectation model is 0.95537,

showing further developed dependability in

recognizing falls.

REFERENCES

M. A. Kuhail, Advances, Applications and the Future of

Haptic Technology. Springer Nature.

T. Yang, X. Lu, L. Yang, M. Yang, J. Chen, and H. Zhao,

“Application of MRI image segmentation algorithm

for brain tumors based on improved YOLO,” Front

Neurosci, vol. 18, p. 1510175, 2024.

H. Jing et al., “Development and Validation of a

Predictive Model for Fall Risk in Pre-Frail Older

Adults,” Res Gerontol Nurs, vol. 18, no. 1, pp. 29–39,

Jan. 2025.

Singh, Navneet Pratap, R. Ravichandran, Soumi Ghosh,

Priya Rana, Shweta Chaku, and Jagendra Singh.

"Enhancing Healthcare Security Using IoT-Enabled

with Continuous Authentication Using Deep

Learning." In International Conference on Electrical

and Electronics Engineering, pp. 275-289. Singapore:

Springer Nature Singapore, 2023.

L. Liu, Y. Sun, Y. Li, and Y. Liu, “A hybrid human fall

detection method based on modified YOLOv8s and

AlphaPose,” Sci Rep, vol. 15, no. 1, p. 2636, Jan.

2025.

C.-W. Kang, Z.-K. Yan, J.-L. Tian, X.-B. Pu, and L.-X.

Fast and High Precision Human Fall Prevention and Intimation Using Haptic Technology

637

Wu, “Constructing a fall risk prediction model for

hospitalized patients using machine learning,” BMC

Public Health, vol. 25, no. 1, p. 242, Jan. 2025.

J. Stuby, P. Leist, N. Hauri, S. Jeevanji, M. Méan, and C.

E. Aubert, “Intervention to systematize fall risk

assessment and prevention in older hospitalized

adults: a mixed methods study,” BMC Geriatr, vol.

25, no. 1, p. 45, Jan. 2025.

AS Kalyana Kumar, et.al, “Artificial Intelligence in

Digital Currency Security: Transforming Global

Marketing in the Blockchain Era”, Cuestiones de

Fisioterapia, Vol.54, No.3, pp. 1907-1928, February

2025.

E. Cubo et al., “Cost-utility analysis of a coadjutant

telemedicine intervention for fall prevention in

Parkinson’s disease,” Eur J Neurol, vol. 32, no. 1, p.

e16561, Jan. 2025.

C. Chen, H. Song, H. Xu, M. Chen, Z. Liang, and M.

Zhang, “Fall risk factors and mitigation strategies for

hematological malignancy patients: insights from a

qualitative study using the reason model,” Support

Care Cancer, vol. 33, no. 2, p. 118, Jan. 2025.

D. Gnutt et al., “Stability Effect of Quinary Interactions

Reversed by Single Point Mutations,” J Am Chem

Soc, vol. 141, no. 11, pp. 4660–4669, Mar. 2019.

Banu, M. Sheerin, E. Baraneetharan, B. Vinothkumar, and

Muruganantham Ponnusamy. "ADVANCEMENT IN

CIRCUIT TECHNOLOGIES FOR ENERGY

HARVESTING SYSTEMS."

T. Ishigaki et al., “Changes in glenohumeral range of

motion by repetitive pitching and their relationship

with arm speed during pitching,” Sports Biomech, pp.

1–13, Jan. 2025.

J. Gudin, M. Sakr, J. Fason, and P. Hurwitz, “Piezo Ion

Channels and Their Association With Haptic

Technology Use: A Narrative Review,” Cureus, vol.

17, no. 1, p. e77433, Jan. 2025.

K. Pacheco-Barrios, J. Ortega-Márquez, and F. Fregni,

“Haptic Technology: Exploring Its Underexplored

Clinical Applications-A Systematic Review,”

Biomedicines, vol. 12, no. 12, Dec. 2024, doi:

10.3390/biomedicines12122802.

Meenakshi Sharma, et.al, “IoT-Embedded Deep Learning

Model for Real-Time Remote Health Monitoring and

Early Identification of Diseases”, IEEE Explorer -3rd

International Conference on Technological

Advancements in Computational Sciences (ICTACS),

2023.

A. Musa, Predictive Technology: Balancing Privacy With

Possibility. Recorded Books, 2024.

J. Chadokiya et al., “Advancing precision cancer

immunotherapy drug development, administration,

and response prediction with AI-enabled Raman

spectroscopy,” Front Immunol, vol. 15, p. 1520860,

2024.

X. Hu, Q. He, H. Ma, J. Li, Y. Jiang, and K. Wang,

“Flexible Eyelid Pressure and Motion Dual-Mode

Sensor Using Electric Breakdown-Induced

Piezoresistivity and Electrical Potential Sensing,”

ACS Appl Mater Interfaces, Jan. 2025, doi:

10.1021/acsami.4c21230.

Singh, Jagendra, Navneet Pratap Singh, B. Vinothkumar,

Nitin Arvind Shelke, Deepak Sharma, and Abbas

Thajeel Rhaif Alsahlanee. "Deep Learning Model for

Predicting Rice Plant Disease Identification and

Classification for Improving the Yield." In

International Conference on Intelligent Systems

Design and Applications, pp. 138-147. Cham: Springer

Nature Switzerland, 2023.

N. T. Newaz and E. Hanada, “An Approach to Fall

Detection Using Statistical Distributions of Thermal

Signatures Obtained by a Stand-Alone Low-

Resolution IR Array Sensor Device,” Sensors (Basel),

vol. 25, no. 2, Jan. 2025, doi: 10.3390/s25020504.

M. Szántó et al., “Developing a Health Support System to

Promote Care for the Elderly,” Sensors (Basel), vol.

25, no. 2, Jan. 2025, doi: 10.3390/s25020455.

O. Korostynska and A. Mason, Advanced Sensors for

Real- Time Monitoring Applications. MDPI, 2021.

Dr.C.Rajan, et.al, “A Deep Learning based Tire Quality

Inspection System”, International Journal of

Advanced Engineering Science and Information

Technology (IJAESIT), ISSN: 2349-3216 Vol.6, No.6,

June 2023.

W. Kemmler, M. Fröhlich, and H. Kleinöder, Whole-body

Electromyostimulation: A Training Technology to

Improve Health and Performance in Humans?

Frontiers Media SA, 2020.

M. Tavakoli, S. Farokh Atashzar, A. L. Trejos, S. DiMaio,

and P. M. Pilarski, Robotics, Autonomous Systems

and AI for Nonurgent/Nonemergent Healthcare

Delivery During and After the COVID-19 Pandemic.

Frontiers Media SA, 2022.

Pradeep, R., B. Vinothkumar, M. Udhayakumar, and S.

Dhanalaksmi. "Comparative Analysis of OOK, BPSK,

DPSK and PPM Modulation Techniques for

Intersatellite Free-Space Optical Communication."

system 6, no. 02 (2019).

A. Bohr and K. Memarzadeh, Artificial Intelligence in

Healthcare. Academic Press, 2020.

Tareq Z. Ahram and Christianne S. Falcão, Human Factors

and Wearable Technologies. AHFE International,

2022.

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

COMMUNICATION, AND COMPUTING TECHNOLOGIES

638