一等獎

AI (Artificial Intelligence) technology has developed very rapidly in recent years, to the point where it can make fake videos or photos called "deepfake''. According to Sumsub Identity Fraud Report 2023 just in the past year, in the APAC region the number of deepfakes has grown 1530%, in the philippines an astounding 4500% and in 馬來西亞 a 1000% increase, these numbers will continue to rise without a proper defense against them, With this rapidly developing technology, there are several threats from misuse deepfake, namely making fraud via video calls, fake videos to blame innocent people, and so on. Therefore, in this research project, an algorithm architecture will be created, namely a system and method used to detect "deepfake" images. The architecture of this algorithm involves convolution functions, neural networks, convolutional neural networks, data normalization functions, namely ReLu and SoftMax, and pooling. This architecture will then be trained over and over with 140,000 scrambled images, which then will make the architecture ready to be used. By researching and combining this algorithm architecture, a system is produced without a cost and with a final result of up to 90% accuracy and detection of 32 images faster than a human can blink.

本研究旨在運用自然語言處理技術，建立辨識惡意程式的模型。首先在資料集方面，分為 PE資料集與 ELF資料集，均包含良性及惡意執行檔，且蒐集惡意程式時採用多種不同之惡意程式家族。接著對資料集進行反組譯及前處理。使用反組譯後的組合語言檔作為文本，訓練模型以區分良性和惡意程式。研究比較詞袋模型、序列模型、BERT 以及不同 n-gram對模型的影響。 研究結果顯示。詞袋模型以使用multi-hot編碼表現最佳，PE資料集之F1-score為96.87%；序列模型則是有位置編碼的 Transformerencoder 表現最優。在不同 n-gram 的比較，multi-hot詞袋模型與 TF-IDF 詞袋模型，分別在 2-gram 及 5-gram 有最高的 F1-score。

星系的恆星形成速率與恆星質量關係一直是天文研究中的熱門主題。為何星系在相似的恆星總質量下有不同的恆星形成速率？這些差異的基本機制是什麼？為了回答這些問題，在這個研究中，我使用了 Sloan Digital Sky Survey(SDSS)的大量光譜數據以及 Hyper Suprime- Cam(HSC)Survey 所獲取的深度成像數據，來研究星系合併對恆星形成速率的影響。首先， 我使用 HSC 圖像來尋找正在合併過程中的星系子集，之後使用了機器學習演算法幫助辨別所有在 SDSS 數據集中有 HSC 觀測的合併星系。然後，我分析合併的不同階段對恆星形成速率與恆星質量關係的影響，並將其與獨立星系的關係分別進行比較。最終，結果顯示，平均而言，合併星系的星形成速率約是恆星質量相似的獨立星系的 2 倍，這證明星系合併是影響星系恆星形成速率與恆星質量關係的關鍵機制之一。

The study aimed to resolve global phosphate scarcity by developing a cost-effective method for phosphorus recovery from industrial wastewater. In existing wastewater treatment, oxidizing phosphite ions (PHO32-) posed a significant challenge. However, our research aimed to develop an effective method for this oxidation process, crucial for phosphorus recovery in industrial wastewater treatment. By utilizing low-cost iron compounds and innovative catalysts, such as iodine obtained from seawater and copper, we achieved remarkable success. Our method demonstrated the ability to oxidize over 80% of PHO32- into phosphate ions (PO43-) within 120 minutes, overcoming the limitations of existing costly methods involving palladium catalysts or high-voltage conditioned ozone (O3). Moreover, the process exhibited profitability, with a gross profit of $1.84 per kilogram of phosphorus, presenting a drastic reduction in cost compared to conventional methods using palladium catalysts. This breakthrough not only offers sustainable wastewater purification but also promises a pathway for resource recovery. Additionally, our future prospects involve refining this method into a device capable of purifying industrial wastewater and recovering phosphorus, emphasizing sustainability and reduced power consumption through innovative techniques like using iron and carbon plates forming a battery. This novel technology represents a sustainable solution utilizing abundant resources such as iodine from seawater, iron, and calcium, paving the way for sustainable phosphorus resource recovery.

腦溢血是由於腦血管的破裂出血所致的嚴重醫療事件，雖能以開顱手術降低原發性腦損傷所致的物理傷害，但尚無特效藥能改善患者手術預後。本研究旨在開發腦溢血治療的新療法。藉由紅血球與微膠細胞（BV-2 Cell Line）的離體實驗模擬腦溢血病患殘留於腦中的血腫塊與微膠細胞在腦部的吞噬情形，探討神經胜肽 Urocortin (UCN)的清除血腫塊的療效。以螢光標籤的方式確認微膠細胞的紅血球吞噬作用，分析 UCN 對微膠細胞吞噬紅血球的量值 （Phagocytosis Index）、及發炎(M1)/抗發炎(M2)的基因變化（RT-qPCR），發現 UCN 能有效增強微膠細胞的吞噬能力，同時亦能調控其 M1/M2 的作用。期望此研究結果能有助於了解 UCN 清除腦血腫塊的作用，作為開發腦溢血新療法的參考依據。

再生醫學透過控制幹細胞分化以應用於修復受損或是建構新組織。目前誘導幹細胞分化的方法多為添加成長因子的化學誘導法，然而此方法具有高成本和費時等缺點。在本篇研究中，我們以人類臍帶間質幹細胞，作為誘導分化對象，使其貼附於具最適化圖形的膠原蛋白微島嶼，結合大氣電漿及 H2O2 二種氧化應激法刺激細胞，探討誘導 WJ-MSC 幹細胞分化的最佳條件。在本研究中，我們發現單獨使用大氣電漿處理可使成骨分化率增加，而大氣電漿處理再加上星形微島嶼培養細胞時，則更顯著的提高成骨分化率。未來嘗試以 H2O2 誘導幹細胞分化， 並進一步探討大氣電漿和 H2O2 提高脂肪分化率的條件、影響分化的機制，期望能此新技術取代傳統化學誘導法，對組織工程做出貢獻。

Understanding species abundance is critical to managing and conserving planetary biodiversity. Pacific salmon (Oncorhynchus spp.) are keystone species of cultural, economic, and ecological importance in Alaska and especially Southwest Alaska. Traditional methods of enumerating salmon such as weirs and visual surveys are often costly, time-intensive, and reliant on taxonomic expertise. Environmental DNA (eDNA), which identifies and quantifies species based on DNA they shed in their habitats, is a potential cost- and time- saving alternative. The relative ease of collecting eDNA samples also enables citizen scientist involvement, expanding research coverage. Currently, more research is required to define eDNA’s potential and limits. This project investigates whether quantitative eDNA metabarcoding can accurately quantify the abundances of six fish species: the five Pacific salmon species plus rainbow trout. Water samples were collected from eight creeks in the Wood River watershed of Southwest Alaska. eDNA metabarcoding and subsequent bioinformatics processing produced a read count for each species. These were compared to visual survey counts, taken to be the true counts for the purposes of this study. Data analyses showed a positive, linear relationship between visual survey count and eDNA count for sockeye salmon. The regressions were significant for both the early (p = 0.089) and late (p = 0.030) sampling dates when 𝛼 = 0.10. eDNA detections of non-sockeye species generally corresponded to visual survey observations of species presence or absence. Overall, the results of this study support eDNA’s potential to be an alternative or supplement to standard methods for the enumeration of fish species.

Substance losses, burns and injuries arising from various causes represent a constant problem encountered by homo sapiens throughout its existence. Rudimentary treatments, relatively effective and less effective, have left their mark on the way we perceive the presence and treatment of wounds caused by various factors. Searching through medical archives, we can trace the specific protocols for these medical conditions back to 2200 BC, when they were structured in three steps: 1) cleaning the skin lesion, 2) applying a dressing (from glue to various preparations) and 3 ) bandaging the skin lesion. Currently, the appearance of wounds on the skin is caused both by accidents in the performance of various activities and by certain diseases that manifest themselves through skin rashes or skin lesions. Their frequency is in the thousands, according to the latest statistics, affecting the majority of the population non-selectively. It is vital that skin lesions receive the necessary care and attention, commensurate with their severity. Being open wounds on the surface of the skin, it is essential that the treatment be meticulous and appropriate to their type, as skin lesions represent a threat to the patient's life. From infections to hydroelectrolytic imbalances specific to burns, the multitude of factors that influence healing highlight the need for a dressing that can be easily customized according to the specificity of the wound, the needs of the patient and that is affordable both from the point of view of production cost as well as its use, making death from skin lesions easily avoidable through an intelligent approach. One of the most complex biological processes and indispensable to humans is the healing of skin lesions. Healing involves a carefully regulated series of biochemical and cellular activities in tandem. Traditional therapies and substances of natural origin have been used to facilitate the regeneration process and accelerate the wound healing process, being applied with encouraging results. Despite the fact that these generally present a low cost, they can be more expensive than contemporary treatments and can be influenced by regional, seasonal factors, showing fluctuations from batch to batch, which could lead to unpredictable allergic reactions, side effects and inconsistent clinical findings. Currently, the standard of care for skin lesions is to clean the wound with antiseptic solutions to prevent infection, apply a dressing followed by bandaging to keep the dressing in place, and if necessary excision of the tissue that has become non-viable. In the case of diabetic ulcers, it is necessary to excise the tissue that has become non-viable and to maintain control over the level of glucose in the body.

Vitas is a combination of binaural beats (BB) with background white noise (ASMR) as accompanying music without reducing the quality of brain stimulation and is able to stimulate the left and right brain while synchronizing brain waves so that it can be used as a media for coping with stress for Gen Z students. To find out the effect Vitas in reducing stress levels, research was conducted on high school students belonging to Gen Z. This research aims to measure the stress index of students belonging to Gen Z, then find the effect of the combination of binaural beats (BB) and ASMR on students'stress levels, and finding the best audio combination to reduce stress levels as a coping mechanism. This research uses a quantitative approach with a quasi-experimental design using a one group pre-test-post-test design scheme. The sample in this study consisted of 105 students who were divided into seven treatment groups to listen to a combination of BB and ASMR with different bandwidths. Stress levels were measured using the DASS 42 stress section questionnaire which has been tested for reliability and validity. The results of the student stress index were 16.26 in the mild stress category. There was a significant difference between the negative group, the alpha + ASMR group and the positive group with a T-Paired test result of less than 0.05. Based on the results of the Post Hoc LSD test, it was found that the combination of binaural beats alpha (12hz) + ASMR was the best combination for reducing stress in students.These findings were developed into a digital therapy application product called "Vitas" as an alternative media for coping with stress for Gen Z.

本文的主要結果有兩部分，第一部分，對於固定的𝑏 ∈ 𝑁以原點 O為觀測點，𝑓(𝑥) = 𝑎𝑥𝑏, 𝑎 ∈ 𝑄為觀測視線，觀測目標為格子點陣列𝑉(𝑚) = {(𝑖, 𝑗)|𝑖, 𝑗 ∈ 𝑁, 1 ≤ 𝑖 ≤ 𝑚, 1 ≤ 𝑗 ≤ 𝑚 }，研究其中可見點的數量與機率。我們發現可見點的數量與歐拉函數及默比烏斯函數有關，可見點的機率也與黎曼𝑧𝑒𝑡𝑎函數具有關聯性。第二部分，對於固定的𝑏 ∈ 𝑁，我們在 𝑥軸與𝑦 軸上布置觀測點，以布置的觀測點為新原點，𝑓(𝑥) = 𝑎𝑥𝑏, 𝑎 ∈ 𝑄為觀測視線，研究將目標點集𝑉(𝑚 × 𝑛) = {(𝑖, 𝑗)|𝑖, 𝑗 ∈ 𝑁, 1 ≤ 𝑖 ≤ 𝑚}完整觀測的布點方法與數量。得到重要成果如下，設正整數𝑚 ≥ 6且𝑇 ⊂ {1, … , 𝑚 + 1}為一個 𝐹(𝑚) −覆蓋，𝑟為大於𝑚的最小質數，對於目標點集𝑉(𝑚 × 𝑛)，建構觀測點集 𝑆2 = {(0, 0), (0, 𝑟)}∪{(𝑡, 0) | 𝑡 ∈ 𝑇}，則 𝑉(𝑚 × 𝑛)為𝑆2 −可見。並進一步研究將目標點集改為𝑉(𝑛 × 𝑚) = {(𝑖, 𝑗)|𝑖, 𝑗 ∈ 𝑁, 1 ≤ 𝑗 ≤ 𝑚}，發現其所需要的觀測點數可顯著減少。