Recently, the Ministry of Industry and Information Technology (MIIT) announced the list of selected candidates for the 2023 Future Industry Innovation Task Leader Program. The brain-computer interface technology company, Zero Unique Thought, from Shanghai's "Da Ling Hao Wan," has successfully made the list in collaboration with Shanghai Jiao Tong University, Shanghai Mental Health Center, the Sixth Hospital of Peking University, and Hangzhou Dianzi University.

At the end of August last year, the MIIT released a list of tasks for the Future Industry Innovation Task Leader Program, mainly focusing on two frontier fields: future manufacturing and future information. It concentrated on four key directions: the metaverse, humanoid robots, brain-computer interfaces, and general artificial intelligence, systematically planning 52 specific tasks. Following the pace of planning, deploying, and implementing in batches, the program aims to promote the innovative development of China's future industries.

Advertisement

In the brain-computer interface industry track, Zero Unique Thought has chosen the niche route of "non-invasive + emotional." Currently, the company's independently developed seven multimodal depression assessment prototypes and the Zero Unique Data Collection System have been deployed in the psychiatric departments of three top-tier hospitals. They have achieved more than 4,500 patient depression status assessments and multimodal data collection from healthy subjects, which will accelerate the application of emotional brain-computer interface technology.

"Take a Picture of Your Mood in 10 Minutes"

Usually, whether it's a fracture or a lung infection, doctors will first ask patients to take an X-ray to check the lesion before making a diagnosis. Can the diagnosis of mental illness also be achieved through a similar "picture-taking" process? Three years ago, Zero Unique Thought, which originated from Shanghai Jiao Tong University, is working hard to turn this idea into reality.

Put on the electroencephalogram (EEG) cap, sit in the seat like playing a racing car game, and answer questions while watching the emotional induction materials displayed on the computer. At Zero Unique Thought, the reporter witnessed the diagnostic process of the "emotional X-ray machine."

"You will see a series of images or videos, please choose the emotion expressed in the picture." With the tester's instructions, an 18th-century oil painting "The Drunken Priestess" appeared on the screen. The protagonist showed a graceful and friendly smile, seemingly slightly intoxicated. At this time, although the subject's face showed no expression, the electroencephalogram curve on the computer screen next to her immediately fluctuated, and her psychological activities were "drawn" in real-time, and the whole process lasted about 10 minutes.

After algorithm processing, a depression status report was completed in 2 minutes. The report clearly marked the quantified risk parameters of the subject's depression and anxiety, and the radar chart displayed several dimensions, which are the main basis for judging depression.

Lu Baoliang, a professor in the Department of Computer Science and Engineering at Shanghai Jiao Tong University and the Chief Scientist of Zero Unique Thought, explained to the reporter that the emotional feedback of patients with depression is different from that of ordinary people. For example, when ordinary people see "The Priestess," they feel happy and peaceful, but patients with depression generally cannot feel happiness, which is because they tend to choose negative options. In the past, when measured by a scale, the subject might hide their true feelings, but it is difficult to escape the "golden eyes" of the "emotional X-ray machine."

The desktop eye movement tracker is also different from general emotional testing equipment. Lu Baoliang said that compared to facial expressions, eye movement signals are more closely related to emotions. Through the eye movement device set below the screen, more than 100 eye movement characteristics such as pupil diameter, blinking frequency, and gaze points can be collected, thus truly reflecting people's emotional state.Pioneering Globally Recognized Emotional Dataset

How are emotions captured? The secret lies within the EEG cap. Its 18 electrodes, like little claws, make contact with the subject's scalp, covering and monitoring all areas of the brain.

The EEG cap is a sensing device, with each "claw" capable of collecting an electroencephalogram (EEG) signal. Lv Baoliang introduced that in the choice of the key technical route for how to collect EEG signals, Zero Unique Thought ultimately chose a non-invasive approach. Although EEG signals collected by non-invasive devices are more difficult to process, they are safer and suitable for a wider range of people.

In addition to continuously improving the wearability of hardware devices such as EEG caps, building a standard dataset for emotional brain-computer interfaces is also an important issue in the field of emotional brain-computer interfaces. Due to the niche research field, expensive collection equipment, and complex data annotation, globally recognized standard emotional EEG datasets are extremely rare.

In 2014, Lv Baoliang's team was the first in the world to propose a multimodal emotional brain-computer interface framework that integrates EEG signals with eye movement signals, and established a dataset for it. Now, the dataset named SEED has become the largest in terms of data volume and the most diverse in terms of data types in the field of emotional EEG datasets internationally, and it is also one of the two most commonly used standard emotional EEG datasets in the field. Since its public release in October 2015, it has been applied for by about 2000 universities and research institutions from 81 countries and regions worldwide, with more than 4880 applications and more than 1400 published papers.

Taking the three types of emotional EEG datasets SEED as an example, Lv Baoliang's team selected 15 movie clips of three emotions as emotional induction materials, and collected and recorded EEG according to international standards. The dataset includes 62-channel EEG and eye movement data for each of the 15 subjects, with each subject having three trials. The results show that eye movement signals are a very good signal for emotion recognition. If EEG signals or eye movement signals are used alone, their respective recognition rates are about 78%. If these two signals are integrated through classic ensemble learning, the recognition rate for the three types of emotions is increased to 88%.

Based on a series of original research in the field of emotional intelligence and emotional brain-computer interfaces, Lv Baoliang was selected as Elsevier's 2023 "Highly Cited Chinese Scholars" list in March this year, which is also the fourth consecutive year since 2020 that he has been selected for this list.

The "Bole" of Minority Innovation Continues

In the past two years, the popularity of brain-computer interfaces has soared. In March this year, the American Neuralink company demonstrated the first case of a brain-computer interface implant patient playing chess and games with their mind. At about the same time, Professor Hong Bo's team from Tsinghua University's School of Medicine helped patients with high-level paraplegia to achieve autonomous brain-controlled drinking through a semi-invasive brain-computer interface.

Facing the "wind" of the future industry, many investors are looking for brain-computer interface technologies and products that represent the future direction in the market. When investors find Zero Unique Thought, they always ask Lv Baoliang a question: "Can your 'Emotion X-ray Machine' get a Class III medical device certificate?"Based on the method of acquiring brain signals through sensors, brain-computer interfaces (BCIs) are divided into invasive and non-invasive types. In the fiercely competitive field of BCIs, many people are still unclear about how non-invasive BCI technology can achieve large-scale application. From the perspective of purpose and function, BCIs can be further divided into two directions: motor and emotional. The relatively niche field of emotions is what Zero Unique Thought has chosen.

Fortunately, on this not-so-popular path of innovation, "patrons" are continuously emerging. In 2019, Lv Baoliang's research on emotional BCIs became part of the first major interdisciplinary research project launched by Shanghai Jiao Tong University, involving medicine and engineering. The project was jointly participated in by seven professors, including Director Sun Bo-Min of the Functional Neurosurgery Department of Ruijin Hospital and Professor Fang Yiru from the Shanghai Mental Health Center. In 2021, Lv Baoliang received investment from Mihoyo to establish Zero Unique Thought, dedicated to the development of the "Emotion X-ray Machine."

Being selected for the Future Industry Innovation Task Unveiling Project has allowed Lv Baoliang to see the spring of "cold research" once again. Data shows that by 2021, the number of psychiatrists in China accounted for less than 1.5% of the total number of doctors in the country. The disproportionate doctor-patient ratio in the field of mental illness highlights the market application prospects of the "Emotion X-ray Machine."

At present, Zero Unique Thought is fully promoting the application for the third-class medical device certificate of the "Emotion X-ray Machine," looking forward to the early benefit of the emotional BCI technology to the public.