<b>Dataset of GPi Deep Brain Stimulation in a Hemi-Parkinsonian rat model</b>

<b>Dataset of GPi Deep Brain Stimulation in a Hemi-Parkinsonian rat model</b><b>Version 0.0</b>Intermittent stimulation influences efficacy of deep brain stimulation on motor symptoms in hemi-Parkinsonian ratsContributed by M. M. Ostad Ali Akbary, A. Dini, and A. Erfanian (2025)<b>Summary</b>In this study, we evaluated the effects of various stimulation parameters on motor symptoms in a hemi-parkinsonian rat model during GPi-DBS. We assessed motor performance across a range of DBS frequencies, amplitudes, and pulse-widths. We also introduced an intermittent stimulation paradigm to reduce the delivered charge compared with continuous stimulation meanwhile improving DBS efficacy.<br><i>A. Animal</i>The experiments were conducted on 16 Wistar rats (male, 300–400 g). The animals were kept in a controlled environment with a 12-hour light‒dark cycle and unrestricted access to food and water. Ethical approval for all animal-related procedures was obtained from the Animal Care and Ethics Committee of the Iran Neural Technology Research Center, which is affiliated with the Iran University of Science and Technology. The study adhered to the recommended guidelines and protocols for the appropriate care and utilization of laboratory animals.B. LesioningThe rats were initially anesthetized via the intraperitoneal injection of a ketamine‒xylazine mixture (100 mg/kg ketamine and 12 mg/kg xylazine). Animal breathing, reflexes, and the level of anesthesia were monitored throughout the duration of the surgery. The depth of anesthesia was assessed by a lack of response to the hind limb or tail pinch. If there was a reaction, then half of the dose was repeated. To generate PD models, a 10 µl solution of 6-hydroxydopamine (6-OHDA) (Sigma Aldrich, GmbH, USA) was prepared in saline at a concentration of 2 mg/ml in 0.01% ascorbic acid and injected into the medial forebrain bundle (MFB, coordinates from bregma: A 2.76 mm, L 1.9 mm, H 8.3 mm) of the rats via a 10 µl microsyringe (80300, Hamilton, United States) with a 26-gauge needle. Two weeks after surgery, the extent of Parkinson's disease in each rat was assessed by injecting a solution of 0.05 mg/kg apomorphine (Sigma Aldrich, GmbH, USA, diluted in 0.2% ascorbic acid). The rats that exhibited a minimum of three anti-clockwise turns per minute were selected as well-lesioned rats.C. Electrode ImplantationThe lesioned rats then underwent stereotactic surgery for electrode implantation while being maintained under a ketamine‒xylazine mixture (100 mg/kg ketamine and 12 mg/kg xylazine). Each rat was implanted with a custom-made, four-electrode stainless steel stimulating microelectrode array into the right entopeduncular nucleus (EP), which corresponds to the globus pallidus internus (GPi, coordinates from bregma: A 2.64 mm, L 2.9 mm, H 7.7 mm). The stimulating microelectrode array was constructed by twisting four wires (impedance of 10 kΩ, diameter of 152 µm, AM system, 793500). To select a pair of electrodes that produced the most effective behavioral response and the fewest side effects during bipolar stimulation, an initial test was performed for each lesion rat. Different pairs of electrode wires were tested during bipolar stimulation, and the effective pair was selected.D. Deep Brain Stimulation DataWe captured video recordings from each rat to analyze the effects of various deep brain stimulation (DBS) parameters, including frequency, amplitude, pulse-width, and intermittent stimulation. These videos were processed with DeepLabCut to track the positions of each rat’s head and back. MATLAB was then used to determine the number of rotations performed during each trial, providing a quantitative measure of the effects of different DBS parameters.<b>Conditions for using the data</b>If you publish any work using this dataset, please cite it in the following recommended format:<br>M. M. Ostad Ali Akbary, A. Dini, and A. Erfanian, “Intermittent Stimulation influences the efficacy of Deep brain stimulation on Motor Symptoms in Hemi-Parkinsonian Rats”.<b>Data files organization</b>The main MATLAB data file (.mat) is organized into five main fields: Frequency, Amplitude, Pulse-Width, Intermittent Stimulation, and Threshold as follows:· <b>Frequency</b>: The Frequency field contains ten subfields, each corresponding to a different evaluated frequency value (with a constant amplitude [I = I₀] and pulse-width [65 µs]). Within each frequency subfield, there are ten individual rat fields, and each rat field contains three trial files (trial 1, trial 2, and trial 3). Every .mat file stores the positional data of the rat’s head and back from a 30-second video recording.· <b>Amplitude</b>: The Amplitude field is organized into five subfields, each representing a distinct evaluated amplitude (with constant frequency [130 Hz] and pulse-width [65 µs]). Each amplitude subfield includes six rat fields, and within each of these, three trial files (trial 1, trial 2, and trial 3) are provided. Each .mat file corresponding to an experimental trial captures the position data of the rat’s head and back from a 30-second video recording.· <b>Pulse-Width</b>: Within the Pulse-Width field, there are five subfields corresponding to different evaluated pulse-width values (maintaining constant frequency [130 Hz] and amplitude [I = I₀]). Each subfield is further divided into six rat fields, and each rat field holds three trial files (trial 1, trial 2, and trial 3). Each .mat file corresponding to an experimental trial captures the position data of the rat’s head and back from a 30-second video recording.· <b>Intermittent Stimulation</b>: The Intermittent Stimulation field is divided into two categories based on the on-time of stimulation. For the 2-second on-time condition, the field includes six subfields representing different evaluated off-time values (with constant frequency [130 Hz], amplitude [I = I₀], and pulse-width [65 µs]). Inside each off-time field, there are eight individual rat fields, each containing three trial files (trial 1, trial 2, and trial 3) with position data from a 30-second video recording. For the 3-second on-time condition, the field comprises three subfields for different evaluated off-time values (under the same constant parameters), where each off-time field contains five rat fields. Each .mat file corresponding to an experimental trial captures the position data of the rat’s head and back from a 30-second video recording.· <b>Threshold</b>: The Threshold field consists of eight subfields, each corresponding to an individual rat. Within these fields, the number of trial files varies according to the number of experiments conducted to achieve the threshold (with constant frequency [130 Hz] and pulse-width [65 µs]). Each .mat file corresponding to an experimental trial captures the position data of the rat’s head and back from a 30-second video recording.<b>Data format</b>The dataset is provided in MATLAB format (.mat). Each .mat file contains a matrix of dimensions <i>n × m</i>, where:· <i>n</i>: Represents the number of frames in the 30-second video recording.· <i>m:</i> Represents the positional data for the rat’s head and back in each frame. Specifically, the matrix contains:(x, y) coordinates for the rat’s head,(x, y) coordinates for the rat’s back,A likelihood value for each head and back position. This value indicates how confident the algorithm is that the labeled point is the true landmark, with values closer to one being more confident.<b>单侧帕金森病大鼠模型的苍白球内侧部(Globus Pallidus Internus, GPi)脑深部电刺激数据集</b><b>版本0.0</b><br>间歇性刺激对单侧帕金森病大鼠脑深部电刺激(Deep Brain Stimulation, DBS)运动症状疗效的影响，由M. M. Ostad Ali Akbary、A. Dini与A. Erfanian贡献（2025年）<br><b>摘要</b>：本研究针对单侧帕金森病大鼠模型的GPi-DBS，评估了不同刺激参数对其运动症状的影响。我们系统测试了不同DBS频率、振幅与脉冲宽度下的运动行为表现，并提出一种间歇性刺激范式，相较于持续刺激，该范式可降低刺激总电荷量，同时提升DBS的治疗疗效。<br><i>A. 实验动物</i>：本实验使用16只雄性Wistar大鼠，体重范围为300~400g。实验动物饲养于环境可控的饲养室，采用12小时明暗循环光照制度，自由进食饮水。所有动物实验操作均已获得隶属于伊朗科技大学的伊朗神经技术研究中心动物护理与伦理委员会的伦理审批，本研究严格遵循实验动物护理与使用的相关规范与指南。<br><i>B. 造模损伤</i>：大鼠初始采用腹腔注射氯胺酮-赛拉嗪混合麻醉剂（100mg/kg氯胺酮+12mg/kg赛拉嗪）进行麻醉。手术全程监测大鼠呼吸、反射及麻醉深度，通过后肢或尾部捏夹无反应来判断麻醉深度，若出现反应则追加半量麻醉剂。为构建帕金森病(Parkinson's Disease, PD)模型，我们将10μl浓度为2mg/ml的6-羟基多巴胺(6-OHDA)（Sigma Aldrich GmbH, 美国）溶液（溶于含0.01%抗坏血酸的生理盐水），通过26号针头的10μl微量注射器（80300，Hamilton，美国）注入大鼠内侧前脑束(Medial Forebrain Bundle, MFB，以前囟为原点的坐标：A=2.76mm，L=1.9mm，H=8.3mm)。术后两周，通过腹腔注射0.05mg/kg阿扑吗啡（Sigma Aldrich GmbH, 美国，溶于0.2%抗坏血酸溶液）评估每只大鼠的PD造模程度，选取每分钟至少逆时针旋转3圈的大鼠作为成功造模个体。<br><i>C. 电极植入</i>：造模成功的大鼠在氯胺酮-赛拉嗪混合麻醉下（100mg/kg氯胺酮+12mg/kg赛拉嗪）接受立体定位手术以植入电极。每只大鼠右侧内苍白球（Entopeduncular Nucleus, EP，对应GPi，以前囟为原点的坐标：A=2.64mm，L=2.9mm，H=7.7mm）被植入定制的四电极不锈钢刺激微电极阵列。该刺激微电极阵列由四根导线绞合制成（阻抗10kΩ，直径152μm，AM system，793500）。为筛选在双极刺激下能产生最优行为反应且副作用最少的电极对，我们对每只造模大鼠进行术前测试：测试不同电极导线配对的双极刺激效果，最终选定最优电极对。<br><i>D. 脑深部电刺激实验数据</i>：我们录制每只大鼠的行为视频，以分析不同DBS参数（包括频率、振幅、脉冲宽度及间歇性刺激）对其运动行为的影响。通过DeepLabCut处理视频，追踪大鼠头部与背部的位置坐标，随后使用MATLAB统计每轮实验中的旋转次数，以此量化不同DBS参数的干预效果。<br><b>数据集使用规范</b>：若您使用本数据集发表研究成果，请按照以下推荐格式引用：<br>M. M. Ostad Ali Akbary、A. Dini与A. Erfanian，《间歇性刺激对单侧帕金森病大鼠脑深部电刺激运动症状疗效的影响》<br><b>数据文件组织形式</b>：主MATLAB数据文件（.mat）包含五个核心字段：频率（Frequency）、振幅（Amplitude）、脉冲宽度（Pulse-Width）、间歇性刺激（Intermittent Stimulation）与阈值（Threshold），具体结构如下：<br>· <b>频率（Frequency）</b>：该字段包含10个子字段，每个子字段对应一个测试频率值（振幅固定为I=I₀，脉冲宽度固定为65微秒）。每个频率子字段下包含10个大鼠个体字段，每个大鼠个体字段包含3轮实验文件（实验1、实验2与实验3）。每一个.mat文件存储了30秒视频录制中大鼠头部与背部的位置数据。<br>· <b>振幅（Amplitude）</b>：该字段包含5个子字段，每个子字段对应一个测试振幅值（频率固定为130Hz，脉冲宽度固定为65微秒）。每个振幅子字段下包含6个大鼠个体字段，每个大鼠个体字段包含3轮实验文件（实验1、实验2与实验3）。每一个对应实验轮次的.mat文件存储了30秒视频录制中大鼠头部与背部的位置数据。<br>· <b>脉冲宽度（Pulse-Width）</b>：该字段包含5个子字段，每个子字段对应一个测试脉冲宽度值（频率固定为130Hz，振幅固定为I=I₀）。每个子字段下包含6个大鼠个体字段，每个大鼠个体字段包含3轮实验文件（实验1、实验2与实验3）。每一个对应实验轮次的.mat文件存储了30秒视频录制中大鼠头部与背部的位置数据。<br>· <b>间歇性刺激（Intermittent Stimulation）</b>：该字段根据刺激开机时长分为两类：① 开机时长2秒条件下：包含6个子字段，每个子字段对应一个测试关机时长值（频率固定为130Hz，振幅固定为I=I₀，脉冲宽度固定为65微秒）。每个关机时长子字段下包含8个大鼠个体字段，每个大鼠个体字段包含3轮实验文件（实验1、实验2与实验3），存储30秒视频录制中的大鼠头部与背部位置数据。② 开机时长3秒条件下：包含3个子字段，每个子字段对应一个测试关机时长值（参数与上述一致），每个关机时长子字段下包含5个大鼠个体字段。每一个对应实验轮次的.mat文件存储了30秒视频录制中大鼠头部与背部的位置数据。<br>· <b>阈值（Threshold）</b>：该字段包含8个子字段，每个子字段对应一只大鼠个体。该字段下的实验文件数量根据阈值测试的实验次数而定（频率固定为130Hz，脉冲宽度固定为65微秒）。每一个对应实验轮次的.mat文件存储了30秒视频录制中大鼠头部与背部的位置数据。<br><b>数据格式</b>：本数据集以MATLAB格式（.mat）提供。每个.mat文件包含一个n×m维度的矩阵，其中：<br>· <i>n</i>：代表30秒视频录制的总帧数。<br>· <i>m</i>：代表单帧中大鼠头部与背部的位置数据。具体而言，该矩阵包含：大鼠头部的(x,y)坐标、大鼠背部的(x,y)坐标、头部与背部每个标记点的置信度值。该数值表示算法对标记点为真实解剖地标置信程度，数值越接近1则置信度越高。