To develop a computer vision-based method for automating the Revised NIOSH Lifting Equation (RNLE) by identifying lift start and end times and estimating key RNLE multipliers from video data.
Background
The RNLE is widely used to assess low back injury risk during lifting tasks; however, its traditional application relies on manual measurements that are labor-intensive and prone to human error. Recent advances in computer vision offer opportunities to automate this process.
Method
The proposed method follows a three-stage process: (1) BlazePose, a real-time pose estimation model, detects 22 key body joints, followed by low-pass filtering to reduce noise. (2) Kinematic features are extracted from these joints, and video frames are classified as lifting or nonlifting to identify lifting phases and lift start and end times. (3) RNLE multipliers are estimated at these lift timings to compute the recommended weight limit (RWL).
Results
The method achieved a mean absolute timing error of 0.25 s for lift timing identification and showed strong correlations between estimated and ground-truth multipliers. The mean absolute error for RWL estimation was 1.58 kg, with a correlation coefficient of 0.91.
Conclusion
These results demonstrate the feasibility of using computer vision to automate the RNLE. Further improvements in lift timing identification, depth estimation, and validation in more diverse workplace settings are recommended.
Application
The proposed method enables RNLE implementation on widely available platforms such as mobile devices and surveillance cameras, promoting safer lifting practices in real-world workplace environments.
AliM. M.Medhat HassanM.ZakiM. (2024). Human pose estimation for clinical analysis of gait pathologies. Bioinformatics and Biology Insights, 18, 11779322241231108. https://doi.org/10.1177/11779322241231108
2.
AnagnostisA.BenosL.TsaopoulosD.TagarakisA.TsolakisN.BochtisD. (2021). Human activity recognition through recurrent neural networks for human–robot interaction in agriculture. Applied Sciences, 11(5), 2188. https://doi.org/10.3390/APP11052188
3.
AngeliniF.FuZ.LongY.ShaoL.NaqviS. M. (2019). 2D pose-based real-time human action recognition with occlusion-handling. IEEE Transactions on Multimedia, 22(6), 1433–1446. https://doi.org/10.1109/tmm.2019.2944745
4.
ArjmandN.AminiM.Shirazi-AdlA.PlamondonA.ParnianpourM. (2015). Revised NIOSH lifting equation may generate spine loads exceeding recommended limits. International Journal of Industrial Ergonomics, 47, 1–8. https://doi.org/10.1016/J.ERGON.2014.09.010
5.
BangaruS. S.WangC.BusamS. A.AghazadehF. (2021). ANN-based automated scaffold builder activity recognition through wearable EMG and IMU sensors. Automation in Construction, 126, 103653. https://doi.org/10.1016/J.AUTCON.2021.103653
6.
\BarimM. S.SesekR. F.CapanogluM. F.DrinkausP.SchallM. C.GallagherS.DavisG. A. (2019). Improving the risk assessment capability of the revised NIOSH lifting equation by incorporating personal characteristics. Applied Ergonomics, 74, 67–73. https://doi.org/10.1016/J.APERGO.2018.08.007
BazarevskyV.GrishchenkoI.RaveendranK.ZhuT.ZhangF.GrundmannM. (2020). BlazePose: On-device real-time body pose tracking. https://arxiv.org/abs/2006.10204v1
9.
BodaS.GargA.Campbell-KyureghyanN. (2012). Can the revised NIOSH lifting equation predict low back pain incidence in a ‘90-day-pain-free-cohort’?Proceedings of the Human Factors and Ergonomics Society Annual Meeting, 56(1), 1178–1182. https://doi.org/10.1177/1071181312561256
10.
CheginiA.GhousiR.NaeiniH. S.KaruppiahK. (2020). Developing a new vertical multiplier to modify the revised NIOSH lifting equation. International Journal of Occupational Hygiene, 12(2), 107–121. https://ijoh.tums.ac.ir/index.php/ijoh/article/view/469
11.
ChenH.LiuP.ZhouG.LuM. L.YuD. (2025). Computer vision and tactile glove: A multimodal model in lifting task risk assessment. Applied Ergonomics, 127, 104513. https://doi.org/10.1016/j.apergo.2025.104513
12.
ColeS. S.McGlothlinJ. (2009). Ergonomics case study: Revised NIOSH lifting equation instruction issues for students. Journal of Occupational and Environmental Hygiene, 6(11), D73–D81. https://doi.org/10.1080/15459620903249695
DempseyP. G.McGorryR. W.MaynardW. S. (2005). A survey of tools and methods used by certified professional ergonomists. Applied Ergonomics, 36(4), 489–503. https://doi.org/10.1016/J.APERGO.2005.01.007
15.
DickR. B.HudockS. D.LuM.-L.WatersT. R.Putz-AndersonV. (2016). Manual materials handling. In Physical and biological hazards of the workplace (pp. 33–52). John Wiley & Sons. https://doi.org/10.1002/9781119276531.CH3
16.
EigenD.PuhrschC.FergusR. (2014). Depth map prediction from a single image using a multi-scale deep network. In Advances in neural information processing systems, 27.
17.
EmodiC. T.ZhangW. J.LangS. Y. T.BiZ. M. (2007). A framework for modeling and analysis of human repetitive operations in a production/assembly line. SAE Technical Papers, 1. https://doi.org/10.4271/2007-01-2500
18.
FanC.MeiQ.YangQ.LiX. (2022). Computer-vision based rapid entire body analysis (REBA) estimation. In Modular and Offsite Construction (MOC) summit proceedings (pp. 90–97). https://doi.org/10.29173/MOCS269
19.
GargA.BanaagJ. (1988). Maximum acceptable weights, heart rates and RPEs for one hour’s repetitive asymmetric lifting. Ergonomics, 31(1), 77–96. https://doi.org/10.1080/00140138808966650
20.
GargA.BodaS.HegmannK. T.MooreJ. S.KapelluschJ. M.BhoyarP.ThieseM. S.MerryweatherA.Deckow-SchaeferG.BloswickD.MalloyE. J. (2014). The NIOSH lifting equation and low-back pain, part 1: Association with low-back pain in the BackWorks prospective cohort study. Human Factors, 56(1), 6–28. https://doi.org/10.1177/0018720813486669
21.
GargA.KapelluschJ. M. (2016). The Cumulative Lifting Index (CULI) for the revised NIOSH lifting equation. Human Factors, 58(5), 683–694. https://doi.org/10.1177/0018720815627405
22.
GolestaniN.MoghaddamM. (2020). Human activity recognition using magnetic induction-based motion signals and deep recurrent neural networks. Nature Communications, 11(1), 1–11. https://doi.org/10.1038/s41467-020-15086-2
23.
GreeneR. L.ChenG.LuM. L.Hen HuY.RadwinR. G. (2021). Enhancing the revised NIOSH lifting equation using ComputerVision. Proceedings of the Human Factors and Ergonomics Society - Annual Meeting, 65(1), 467–471. https://doi.org/10.1177/1071181321651211
24.
GreeneR. L.LuM. L.BarimM. S.WangX.HaydenM.HuY. H.RadwinR. G. (2022). Estimating trunk angle kinematics during lifting using a computationally efficient computer vision method. Human Factors, 64(3), 482–498. https://doi.org/10.1177/0018720820958840
25.
HortonL. M.NussbaumM. A.AgnewM. J. (2015). Rotation during lifting tasks: Effects of rotation frequency and task order on localized muscle fatigue and performance. Journal of Occupational and Environmental Hygiene, 12(2), 95–106. https://doi.org/10.1080/15459624.2014.957829
26.
JungS.SuB.LuL.QingL.XuX. (2025). Video-based lifting action recognition using rank-altered kinematic feature pairs. Human Factors, 67(7), 656–672. https://doi.org/10.1177/00187208241309748
27.
JungS.SuB.WangH.LuL.XieZ.XuX.FittsE. P. (2022). A computer vision-based lifting task recognition method. Proceedings of the Human Factors and Ergonomics Society - Annual Meeting, 66(1), 1210–1214. https://doi.org/10.1177/1071181322661507
28.
KeyserlingW. M. (2000). Workplace risk factors and occupational musculoskeletal disorders, part 1: A review of biomechanical and psychophysical research on risk factors associated with low-back pain. AIHAJ - American Industrial Hygiene Association, 61(1), 39–50. https://doi.org/10.1080/15298660008984513
29.
KimW.SungJ.SaakesD.HuangC.XiongS. (2021). Ergonomic postural assessment using a new open-source human pose estimation technology (OpenPose). International Journal of Industrial Ergonomics, 84, 103164. https://doi.org/10.1016/J.ERGON.2021.103164
30.
KingmaI.FaberG. S.van DieënJ. H. (2010). How to lift a box that is too large to fit between the knees. Ergonomics, 53(10), 1228–1238. https://doi.org/10.1080/00140139.2010.512983
31.
KingmaI.FaberG. S.van DieënJ. H. (2016). Supporting the upper body with the hand on the thigh reduces back loading during lifting. Journal of Biomechanics, 49(6), 881–889. https://doi.org/10.1016/J.JBIOMECH.2015.09.035
32.
LiL.MartinT.XuX. (2020). A novel vision-based real-time method for evaluating postural risk factors associated with musculoskeletal disorders. Applied Ergonomics, 87, 103138. https://doi.org/10.1016/J.APERGO.2020.103138
33.
LiuM.HongD.HanS.LeeS. (2016, May). Silhouette-based on-site human action recognition in single-view video. In Construction research congress 2016 (pp. 951–959).
34.
LouZ.ZhanZ.XuH.LiY.HuY. H.LuM. L.WerrenD. M.RadwinR. G. (2025). A single-camera method for estimating lift asymmetry angles using deep learning computer vision algorithms. IEEE Transactions on Human-Machine Systems, 55(2), 309–314. https://doi.org/10.1109/THMS.2025.3539187
35.
LuM. L.Putz-AndersonV.GargA.DavisK. G. (2016). Evaluation of the impact of the revised national institute for occupational safety and health lifting equation. Human Factors, 58(5), 667–682. https://doi.org/10.1177/0018720815623894
36.
LuM. L.WatersT.WerrenD.PiacitelliL. (2011). Human posture simulation to assess cumulative spinal load due to manual lifting. Part II: Accuracy and precision. Theoretical Issues in Ergonomics Science, 12(2), 189–203. https://doi.org/10.1080/14639220903490740
37.
ManghisiV. M.UvaA. E.FiorentinoM.BevilacquaV.TrottaG. F.MonnoG. (2017). Real time RULA assessment using Kinect v2 sensor. Applied Ergonomics, 65, 481–491. https://doi.org/10.1016/J.APERGO.2017.02.015
38.
MarrasW. S.FineL. J.FergusonS. A.WatersT. R. (1999). The effectiveness of commonly used lifting assessment methods to identify industrial jobs associated with elevated risk of low-back disorders. Ergonomics, 42(1), 229–245. https://doi.org/10.1080/001401399185919
39.
MitalA.PennathurA. (1999). Musculoskeletal overexertion injuries in the United States: Mitigating the problem through ergonomics and engineering interventions. Journal of Occupational Rehabilitation, 9(2), 115–149. https://doi.org/10.1023/A:1021318204926/METRICS
40.
MosabbebE. A.RaahemifarK.FathyM. (2013). Multi-view human activity recognition in distributed camera sensor networks. Sensors, 13(7), 8750–8770. https://doi.org/10.3390/s130708750
41.
NadriH.Fasih RamandiF. (2016). Assessment of manual material handling in a tile and ceramic factory using the national institute for occupational safety and health equation in 2016. Journal of Occupational Health and Epidemiology, 5(2), 105–111. https://doi.org/10.18869/ACADPUB.JOHE.5.2.105
42.
OrdóñezF. J.RoggenD.LiuY.XiaoW.ChaoH.-C.ChuP. (2016). Deep convolutional and LSTM recurrent neural networks for multimodal wearable activity recognition. Sensors, 16(1), 115. https://doi.org/10.3390/S16010115
43.
ParkK. W.MirianM. S.McKeownM. J. (2024). Artificial intelligence-based video monitoring of movement disorders in the elderly: A review on current and future landscapes. Singapore Medical Journal, 65(3), 141–149. https://doi.org/10.4103/SINGAPOREMEDJ.SMJ-2023-189
44.
PizatellaT. J.Putz-AndersonV.BobickT. G.McGlothlinJ. D.WatersT. R. (1992). Understanding and evaluating manual handling injuries: NIOSH research studies. Ergonomics, 35(9), 945–953. https://doi.org/10.1080/00140139208967374
45.
PriscoG.DonisiL.GuerriniL.MercaldoF.EspositoF.SantoneA.CesarelliM.AmatoF.GargiuloP. (2024). Biomechanical risk evaluation through machine learning algorithms fed with features extracted from sEMG of neck extensors. In 2024 IEEE international conference on metrology for Extended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE) (pp. 171–176). IEEE. https://doi.org/10.1109/METROXRAINE62247.2024.10796263
46.
RodriguesP. B.XiaoY.FukumuraY. E.AwadaM.AryalA.Becerik-GerberB.LucasG.RollS. C. (2022). Ergonomic assessment of office worker postures using 3D automated joint angle assessment. Advanced Engineering Informatics, 52, 101596. https://doi.org/10.1016/J.AEI.2022.101596
47.
SaxenaA.SunM.NgA. Y. (2009). Make3D: Learning 3D scene structure from a single still image. IEEE Transactions on Pattern Analysis and Machine Intelligence, 31(5), 824–840. https://doi.org/10.1109/TPAMI.2008.132
48.
ShahuR. (2016). The NIOSH lifting equation for manual lifting and its applications. Journal of Ergonomics, 06(02), 1000159. https://doi.org/10.4172/2165-7556.1000159
49.
SharmaV.GuptaM.KumarA.MishraD. (2021). Video processing using deep learning techniques: A systematic literature review. IEEE Access, 9, 139489–139507. https://doi.org/10.1109/ACCESS.2021.3118541
50.
ThomasB.LuM. L.JhaR.BertrandJ. (2022). Machine learning for detection and risk assessment of lifting action. IEEE Transactions on Human-Machine Systems, 52(6), 1196–1204. https://doi.org/10.1109/THMS.2022.3212666
U.S. Bureau of Labor Statistics. (2020). Occupational injuries and illnesses resulting in musculoskeletal disorders (MSDs). Retrieved May 16, 2025, from. https://www.bls.gov/iif/factsheets/msds.htm
53.
Van Der BeekA. J.MathiassenS. E.WindhorstJ.BurdorfA. (2005). An evaluation of methods assessing the physical demands of manual lifting in scaffolding. Applied Ergonomics, 36(2), 213–222. https://doi.org/10.1016/J.APERGO.2004.10.012
WangH.XieZ.LuL.LiL.XuX. (2021a). A computer-vision method to estimate joint angles and L5/S1 moments during lifting tasks through a single camera. Journal of Biomechanics, 129, 110860. https://doi.org/10.1016/J.JBIOMECH.2021.110860
56.
WangJ.TanS.ZhenX.XuS.ZhengF.HeZ.ShaoL. (2021b). Deep 3D human pose estimation: A review. Computer Vision and Image Understanding, 210, 103225. https://doi.org/10.1016/J.CVIU.2021.103225
57.
WangX.HuY. H.LuM. L. J.RadwinR. (2021c). Load asymmetry angle estimation using multiple-view videos. IEEE Transactions on Human-Machine Systems, 51(6), 734–739. https://doi.org/10.1109/THMS.2021.3112962
58.
WatersT. R.BaronS. L.KemmlertK. (1998). Accuracy of measurements for the revised NIOSH lifting equation. Applied Ergonomics, 29(6), 433–438. https://doi.org/10.1016/S0003-6870(98)00015-5
59.
WatersT. R.LuM. L.PiacitelliL. A.WerrenD.DeddensJ. A. (2011). Efficacy of the revised NIOSH lifting equation to predict risk of low back pain due to manual lifting: Expanded cross-sectional analysis. Journal of Occupational and Environmental Medicine, 53(9), 1061–1067. https://doi.org/10.1097/JOM.0B013E31822CFE5E
WatersT. R.Putz-AndersonV.GargA.FineL. J. (1993). Revised NIOSH equation for the design and evaluation of manual lifting tasks. Ergonomics, 36(7), 749–776. https://doi.org/10.1080/00140139308967940
62.
ZhouG.AggarwalV.YinM.YuD. (2022). A computer vision approach for estimating lifting load contributors to injury risk. IEEE Transactions on Human-Machine Systems, 52(2), 207–219. https://doi.org/10.1109/thms.2022.3148339
63.
ZhuH.WangX.LingX.ZhuY.ZhaoY.LuH. (2024, December). Push up counter based on YOLOV8 deep learning. In 2024 5th international conference on Computers and Artificial Intelligence Technology (CAIT) (pp. 202–206). IEEE.
