Prediction of hazards in laboratory work using deep learning models learnt from past laboratory accidents

Abstract

Laboratory-based research activities frequently involve hazardous materials and operations, which are prone to accidents or injuries. While risk assessment is a necessary step in the research plan, it is often difficult to recognize all potential hazards involved in laboratory work, including those in the preparation and clean-up phases. In this study, we investigate the performance of deep learning models in predicting potential hazards in laboratory work. As a training dataset, actual laboratory accident reports collected from national universities in Japan are labeled with the most suitable hazards. The trained models read a text of laboratory work plan as an input and predict the possible primary and secondary hazards as outputs. The model that combines Bidirectional Encoder Representations from Transformers (BERT) with Bidirectional Long Short-Term Memory (BiLSTM) shows higher performance compared to that of its constituent models individually. The trained models can therefore potentially be used as a core component of risk assessment tool and safety training, enabling junior researchers and students to recognize potential hazards and assess possible accident risks in their laboratory work, thus reducing the frequency of accidents and injuries.

Acknowledgments

This research was partly supported by Grants-in-Aid for Scientific Research from the Ministry of Education, Science, Sports, and Culture of Japan (No. 20K22185).

References

• 1)  Faller G., Mikolajczyk R. T., Akmatov M. K., Meier S., Krämer A.: Accidents in the context of study among university students - A multicentre cross-sectional study in North Rhine-Westphalia, Germany, Accident Analysis & Prevention 42(2), 487-491, 2010.
• 2)  Nishikimi T., Tomita K., Hayashi R., Murata S.: Injury risk estimation for students involved in experimental work in Nagoya University, Journal of Environment and Safety 10(1), 1-7, 2019.
• 3)  Heinrich H. W.: Industrial Accident Prevention, A Scientific Approach, Second., New York & London, McGraw-Hill Book Company Inc., 1941.
• 4)  Benderly B. L.: Report on UH lab explosion reveals deep, systemic safety failures, Science, 2016.
• 5)  Hawkins F. H., Yoshimi I.: Human Factors in Flight, 1st ed., Seizando Shoten, Tokyo, 1992.
• 6)  Song W., Li J., Li H., Ming X.: Human factors risk assessment: An intergrated method for improving safety in clinical use of medical devices, Applied Soft Computing 86, 105918, 2020.
• 7)  Hayashi R., Yamada T., Shinkawa K., Tomita K., Nishikimi T., Murata S., Kurimoto H.: Case study of text analytics applied to accident reports of a university, MATEC Web Conf. 333, 10003, 2021.
• 8)  Murata S., Harada T., Hayashi R., Mishina T., Tomita K.: Eleven risk elements rationalizing laboratory accidents at universities, Journal of Environment and Safety 13(2), 37-45, 2021.
• 9)  Sato Y., Okamoto S., Kayaba K., Nobuhara H., Soeda K.: Effectiveness of role-play in hazard prediction training for nursing students: A randomized controlled trial, Journal of Nursing Education and Practice 8, 1-7, 2018.
• 10)  Hashida M., Kamezaki R., Goto M., Shiraishi J.: Quantification of hazard prediction ability at hazard prediction training (Kiken- Yochi Training: KYT) by free-response receiver-operating characteristic (FROC) analysis, Radiological Physics and Technology 10(1), 106-112, 2017.
• 11)  Senkbeil E. G.: Laboratory safety course in the chemistry curriculum, Journal of Hazardous Materials 36(2), 159-164, 1994.
• 12)  Alaimo P. J., Langenhan J. M., Tanner M. J., Ferrenberg S. M.: Safety teams: An approach to engage students in laboratory safety, Journal of Chemical Education 87(8), 856-861, 2010.
• 13)  Harada T., Moriyama F., Tanaka T.: Improving the chemical management for non-Japanese researchers toward internationalization of university in Japan, Journal of Environment and Safety 11(2), 41-49, 2020.
• 14)  Ménard A. D., Trant J. F.: A review and critique of academic lab safety research, Nature Chemistry 12(1), 17-25, 2020.
• 15)  Ayi H.-R., Hon C.-Y.: Safety culture and safety compliance in academic laboratories: A Canadian perspective, Journal of Chemical Health & Safety 25(6), 6-12, 2018.
• 16)  Yamamoto H.: How Can We Achieve a Practical Way of Risk Assessment of Chemicals in Laboratory?, Journal of Environment and Safety 6(3), 165-168, 2015.
• 17)  Ophir Y., Tikochinski R., Asterhan C. S. C., Sisso I., Reichart R.: Deep neural networks detect suicide risk from textual facebook posts, Scientific Reports 10(1), 16685, 2020.
• 18)  Zhang X., Srinivasan P., Mahadevan S.: Sequential deep learning from NTSB reports for aviation safety prognosis, Safety Science 142, 105390, 2021.
• 19)  Tasdelen A., Sen B.: A hybrid CNN-LSTM model for premiRNA classification, Scientific Reports 11(1), 14125, 2021.
• 20)  Zhang L., Wang S., Liu B.: Deep learning for sentiment analysis: A survey, Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery 8(4), 2018.
• 21)  Wu S., Roberts K., Datta S., Du J. C., Ji Z. C., Si Y. Q., Soni S., Wang Q., Wei Q., Xiang Y., Zhao B., Xu H.: Deep learning in clinical natural language processing: a methodical review, Journal of the American Medical Informatics Association 27(3), 457-470, 2020.
• 22)  Dahl G. E., Yu D., Deng L., Acero A.: Context-Dependent Pre- Trained Deep Neural Networks for Large-Vocabulary Speech Recognition, IEEE Transactions on Audio, Speech, and Language Processing 20(1), 30-42, 2012.
• 23)  Hinton G., Deng L., Yu D., Dahl G. E., Mohamed A., Jaitly N., Senior A., Vanhoucke V., Nguyen P., Sainath T. N., Kingsbury B.: Deep neural networks for acoustic modeling in speech recognition: The shared views of four research groups, IEEE Signal Processing Magazine 29(6), 82-97, 2012.
• 24)  Tian Y., Yang G., Wang Z., Wang H., Li E., Liang Z.: Apple detection during different growth stages in orchards using the improved YOLO-V3 model, Computers and Electronics in Agriculture 157, 417-426, 2019.
• 25)  Ren S. Q., He K. M., Girshick R., Sun J.: Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks, Advances in Neural Information Processing Systems 28 (NIPS 2015), 2015.
• 26)  Wei J., Zou K.: EDA: Easy data augmentation techniques for boosting performance on text classification tasks, arXiv, 2019.
• 27)  Shorten C., Khoshgoftaar T. M., Furht B.: Text Data Augmentation for Deep Learning, Journal of Big Data 8(1), 101, 2021.
• 28)  Devlin J., Chang M.-W., Lee K., Toutanova K.: BERT: Pretraining of deep bidirectional transformers for language understanding, arXiv, 2018.
• 29)  Acheampong F. A., Nunoo-Mensah H., Chen W.: Transformer models for text-based emotion detection: a review of BERTbased approaches, Artificial Intelligence Review 54(8), 5789- 5829, 2021.
• 30)  Joulin A., Grave E., Bojanowski P., Mikolov T.: Bag of tricks for efficient text classification, arXiv, 2016.
• 31)  Bojanowski P., Grave E., Joulin A., Mikolov T.: Enriching word vectors with subword information, arXiv, 2016.
• 32)  Bojanowski P., Grave E., Joulin A., Mikolov T.: Enriching Word Vectors with Subword Information, Transactions of the Association for Computational Linguistics 5, 135-146, 2017.
• 33)  Hochreiter S., Schmidhuber, J.: Long short-term memory, Neural Computation 9(8), 1735-1780, 1997.
• 34)  Graves A., Schmidhuber J.: Framewise phoneme classification with bidirectional LSTM and other neural network architectures, Neural Networks 18(5), 602-610, 2005.
• 35)  Murata S., Tomita K., Mishina T, Hayashi R., Nishikimi T.: Key points regarding safety education related to fire and/or explosion risk control in academic laboratories, Journal of Environment and Safety 10(2), 59-65, 2019.
• 36)  Richards J. M., Sydiskis R. J., Davidson W. M., Josell S. D., Dent Sci. M., Lavine D. S.: Permeability of latex gloves after contact with dental materials, American Journal of Orthodontics and Dentofacial Orthopedics 104(3), 224-229, 1993.
• 37)  Microflex Corporation: Chemical resistance guide, https://betastatic.fishersci.com/content/dam/fishersci/en_US/documents/ programs/healthcare/brochures-and-catalogs/guides/microflexchemical-resistance-guide.pdf, Last accessed at 21 April 2022.
• 38)  Cai R., Qin B., Chen Y., Zhang L., Yang R., Chen S., Wang W.: Sentiment Analysis About Investors and Consumers in Energy Market Based on BERT-BiLSTM, IEEE Access 8, 171408- 171415, 2020.
• 39)  Vlad G.-A., Tanase M.-A., Onose C., Cercel D.-C.: Sentencelevel propaganda detection in news articles with transfer learning and BERT-BiLSTM-Capsule model, Proceedings of the Second Workshop on Natural Language Processing for Internet Freedom: Censorship, Disinformation, and Propaganda, 148-154, Hong Kong, China, Association for Computational Linguistics, 2019.
• 40)  Al-Omari H., Abdullah M. A., Shaikh S.: EmoDet2: Emotion detection in English textual dialogue using BERT and BiLSTM models, 2020 11th International Conference on Information and Communication Systems (ICICS), 226-232, 2020.