3156

Brain network evaluated by functional-guided reinforcement learning effective connectivity indicates theraputic effect in tinnitus patients
Han Lv1
1Radiology, Beijing Friendship Hospital, Capital Medical University, Beijing, China

Synopsis

Keywords: Functional Connectivity, fMRI (resting state), Tinnitus, sound therapy, fMRI, brain network, effective connectivity

Motivation: We hypothesize that the investigation of brain FC may serve as a foundational step in exploring EC.

Goal(s): To enhance the precision of identifying effective connectivity between distinct brain regions.

Approach: We adopts the actor-critic framework with an encoder-decoder model as the actor network. The encoder utilizes the Transformer model, and the decoder employs a bidirectional long-short-term memory (LSTM) network with attention.

Results: Tinnitus patients were feature by enhanced information output of the motor network, somatosensory network, and the visual network. Effective treatment is featured by a similar pattern of FGEC network between tinnitus patients after treatment and the healthy controls.

Impact: The pattern of FGEC network can be regarded as a direct evidence to indicate the effectiveness of sound therapy for tinnitus patients prior to treatment.

Background

Using functional connectivity (FC) or effective connectivity (EC) alone can not effectively delineate the brain network based on functional magnetic resonance imaging (fMRI) data, limited the understanding of the mechanism of tinnitus and its treatment effect.

Methods

We hypothesize that the investigation of brain FC may serve as a foundational step in exploring EC. To enhance the precision of identifying effective connectivity between distinct brain regions, we adopts the actor-critic framework with an encoder-decoder model as the actor network. The encoder utilizes the Transformer model, and the decoder employs a bidirectional long-short-term memory (LSTM) network with attention. We constructed the FGEC network for enrolled subjects per fMRI scan, including 65 tinnitus patients and 28 healthy controls at the time enrollment. After 6 months of sound therapy for tinnitus patients and perspective follow-up, we acquired the fMRI data again and retrospectively categorized the patients as effective group (EG) and ineffective group (IG) according to the treatment effect.

Results

Compared with FC and EC, the FGECRL method demonstrated better accuracy in discriminating different group of subjects, highlighted the advantage of FGECRL in better identifying the features of brain network. For the FGEC network of EG and IG per state (before and after treatment) and healthy controls, tinnitus patients were feature by enhanced information output of the motor network, somatosensory network, and the visual network. Effective treatment is featured by a similar pattern of FGEC network between tinnitus patients after treatment and the healthy controls. Deactivated information output in precentral gyrus, paracentral lobule, postcentral gyrus, and medioventral occipital cortex were biological indicators for effective treatment of tinnitus. Mild to moderate information output of these brain regions were indicators for poor treatment effect, which may indicate abnormality of the whole brain network rather than regional alterations.

Conclusions

The pattern of FGEC network can be regarded as a direct evidence to indicate the effectiveness of sound therapy for tinnitus patients prior to treatment.

Acknowledgements

This study was partially supported by the National Natural Science Foundation of China (62171297, 61931013). The authors would like to thank all the involved study investigators, clinicians, nurses, and technicians for dedicating their time and skills to the completion of this study.

References

1. Y.-F. Cheng, S. Xirasagar, N.-W. Kuo, and H.-C. Lin, “Tinnitus and risk of attempted suicide: A one year follow-up study,” Journal of affective disorders, vol. 322, pp. 141–145, 2023

2. M. E. Adams, T. C. Huang, S. Nagarajan, and S. W. Cheung, “Tinnitus neuroimaging,” Otolaryngologic Clinics of North America, vol. 53, no. 4, pp. 583–603, 2020

3. X. Yu, B. Gong, H. Yang, Z. Wang, G. Qi, J. Sun, Y. Fang, and X. Fan, “Effect of acupuncture treatment on cortical activation in patients with tinnitus: A functional near-infrared spectroscopy study,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 31, pp. 729–737, 2023

4. H. Lv, J. Liu, Q. Chen, Z. Zhang, Z. Wang, S. Gong, J. Ji, and Z. Wang, “Brain effective connectivity analysis facilitates the treatment outcome expectation of sound therapy in patients with tinnitus,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 31, pp. 1158–1166, 2023

Figures

Figure 1. The architecture of the proposed method FGECRL

Figure 2. Encoder architecture

Figure 3. Decoder architecture

Figure 4. The learned FGEC network for each group

Figure 5. Effective connectivity for different groups

Proc. Intl. Soc. Mag. Reson. Med. 32 (2024)
3156
DOI: https://doi.org/10.58530/2024/3156