Google Colab Research Articles

Implementation of Business Intelligence for Product Optimization to Enhance Profitability Through theUtilization of Synergies Among the Product Departments of the Look Company for the Period 2019-2022

This research discusses the implementation of Business Intelligence (BI) to enhance the profitability of The Look Company through the synergy of product departments. In the era of big data, a company's ability to process and analyse data becomes a crucial factor in achieving competitive advantage. This research aims to present data information from The Look Company's Product Departments in the form of visualizations, graphs, and diagrams that can be used as indicators for decision-making. It also aims to identify products that are not generating significant profitability for the company in order to optimize production costs and to identify products that experience frequent cancellations and returns. This analysis and recommendations are intended to assist The Look Company in its decision-making processthrough the implementation of a comprehensive and detailed Business Intelligence study. Additionally, this research outlines the steps of data collection and processing methods using tools such as Google Colab, Postgree, and Tableau to generate visualizations, graphs, and diagrams that provide indicators and recommendations for decision-making within The Look Company. The research results indicate that the BI data visualization dashboard plays a crucial role in determining business strategies and providing rapid and accurate decision-making recommendations.

Hybrid EEG data analysis for diagnosis of stress-related neurological disorder: SKY as an alternative therapy

There are many reasons associated with stress, long term stress induces neurological and psychosomatic disorders like hypertension, hypothyroidism, diabetes, anxiety and depression which affect the lifestyle of human beings. Consequently, behavioural activity and action gradually change in their surrounding environment and also perceived by others. In general, stressful respiration is relatively different from normal. To release stress and control all the neuropsychological hormones, multiple activities like playing games, watching a movie, listening to songs and music, etc. or intake of medicine/drugs such as (Allopathic /Homeopathic/Ayurvedic) are used. Medicines can provide easy stress evasion, but relief is only temporary. Thus, yoga and Sudarshan kriya (SK) meditation is a unique and alternate therapy identified by Gurudev Sri Sri Ravi Shankar by Art of living. It would be a healthy way to get rid of stress in peoples’ lives. Study of long-term effects of (SKY) Sudarshan kriya Yoga before and after and response of the brain regions in experienced (10–15 yrs) practitioners, mediocre (3–5 yrs) and novice(non-practitioners) is the main objective of this work. This study is planned in three phases, the first phase is an experiment on SKY practitioners for more than 10–15 years, in which their (EEG) Electroencephalogram is recorded just after a session of meditation and the common portion of excitation amongst the three subjects is mined and analysed, to draw inferences. This inference would help us draw a conclusion about (BLOC) base level of consciousness considered as benchmark. In the second phase, comparison of benchmark data with the Mediocre (3–5 yrs) measurement and in third phase, benchmark versus Novice data, is done. Next is the phase of interpretation of the response in the form of EEG spectral waves as Type I- 10 to 15 years SKY Practitioners (Superconscious), Type II- SKY practitioners 3 to 5 years (mediocre/semiconscious) and Type. III- Non-practitioner subjects (Novice/Un-conscious). The unconsciousness here means a state of complete unawareness of the self, though conscious of the external, physical world. Thus, power spectrum analysis (PSA) is carried out and frequency of each electrode is computed through segment analysis, Power Spectrum Density (PSD), Correlation coefficient, Mean and Standard Deviation, for finding the level of consciousness. The spectral waveform of these recordings is analysed programmatically using machine learning techniques (used Python Language run on the Jupyter notebook, Spyder, Google colab environment).Frequency analysis results are obtained by placing 21 electrodes in human brain in different lobes that is (Fz, C2, P2, FP1, FP2, F3, F4, C3, C4, P3, P4, O1, O2, F7, F8, T3, T4, T5, BP4, E.G, T6) those are frequency measuring electrodes/channels placed on the frontal lobe, temporal lobe, parietal lobe and occipital lobe over skull and brainwaves alpha (α)[8–12 Hz], beta (β)[12–16 Hz], delta (δ)[0.5–4 Hz], theta (Θ)[4–8 Hz], gamma (γ)[16–32 Hz] are synthesized. The interpretation of these analyses suggests alternative therapeutic techniques, to improve both mentally and psychologically and thus become socially acceptable.

Effect of Hyperparameter Tuning on the Performance of YOLOv8 for Multi Crop Classification on UAV Images

This study investigates the performance of YOLOv8, a Convolutional Neural Network (CNN) architecture, for multi-crop classification in a mixed farm with Unmanned Aerial Vehicle (UAV) imageries. Emphasizing hyperparameter optimization, specifically batch size, the study’s primary objective is to refine the model’s batch size for improved accuracy and efficiency in crop detection and classification. Using the Google Colaboratory platform, the YOLOv8 model was trained over various batch sizes (10, 20, 30, 40, 50, 60, 70, 80, and 90) to automatically identify the five different classes (sugarcane, banana trees, spinach, pepper, and weeds) present on the UAV images. The performance of the model was assessed using classification accuracy, precision, and recall with the aim of identifying the optimal batch size. The results indicate a substantial improvement in classifier performance from batch sizes of 10 up to 60, while significant dips and peaks were recorded at batch sizes 70 to 90. Based on the analysis of the obtained results, Batch size 60 emerged with the best overall performance for automatic crop detection and classification. Although the F1 score was moderate, the combination of high accuracy, precision, and recall makes it the most balanced option. However, Batch Size 80 also shows very high precision (98%) and balanced recall (84%), which is suitable if the primary focus is on achieving high precision. The findings demonstrate the robustness of YOLOv8 for automatic crop identification and classification in a mixed crop farm while highlighting the significant impact of tuning to the appropriate batch size on the model’s overall performance.

Development of a ChatBot model for health telecare: Integration of LangChain, embeddings with OpenAI, and Pinecone using the question answering technique

This study was dedicated to the development of an advanced model for a medical ChatBot, aimed at optimizing Telecare in healthcare. Cutting-edge technologies such as LangChain, Embeddings with OpenAI, and Pinecone were incorporated, applying Question Answering techniques to enhance user-Bot interaction. Google Colab was essential, serving as the main environment for both running the code and the ChatBot, in addition to facilitating intensive data processing. The methodology involved a manual and rigorous data collection from official medical sources to ensure the accuracy and reliability of the information integrated into the ChatBot. The results showed a notable variability in response times between GPT-3.5-turbo and GPT-4, with a comprehensive cost evaluation to analyze the profitability of both models in specific medical contexts. In conclusion, the study provided significant insights into the implementation and efficacy of advanced medical ChatBot models, revealing crucial considerations about optimizing response times and cost evaluations in the development of innovative Telecare solutions in healthcare.

A Versatile Pipeline for Analyzing Dynamic Changes in Nuclear Bodies in a Variety of Cell Types.

Various nuclear processes, such as transcriptional control, occur within discrete structures known as foci that are discernable through the immunofluorescence technique. Investigating the dynamics of these foci under diverse cellular conditions via microscopy yields valuable insights into the molecular mechanisms governing cellular identity and functions. However, performing immunofluorescence assays across different cell types and assessing alterations in the assembly, diffusion, and distribution of these foci present numerous challenges. These challenges encompass complexities in sample preparation, determination of parameters for analyzing imaging data, and management of substantial data volumes. Moreover, existing imaging workflows are often tailored for proficient users, thereby limiting accessibility to a broader audience. In this study, we introduce an optimized immunofluorescence protocol tailored for investigating nuclear proteins in different human primary T cell types that can be customized for any protein of interest and cell type. Furthermore, we present a method for unbiasedly quantifying protein staining, whether they form distinct foci or exhibit a diffuse nuclear distribution. Our proposed method offers a comprehensive guide, from cellular staining to analysis, leveraging a semi-automated pipeline developed in Jython and executable in Fiji. Furthermore, we provide a user-friendly Python script to streamline data management, publicly accessible on a Google Colab notebook. Our approach has demonstrated efficacy in yielding highly informative immunofluorescence analyses for proteins with diverse patterns of nuclear organization across differentcontexts.

PENERAPAN DATA MINING JUMLAH PENJUALAN SEPEDA MOTOR MENGGUNAKAN METODE K- MEANS CLUSTERING (Studi Kasus : PT. Tunas Dwipa Matra Way Kanan)

With the ever-expanding product spectrum of motorcycles, as we all know, motorcycles are a popular form of public transportation because of their practicality. The motorcycle business had to develop further and adopt new technologies. The motorcycle market in Indonesia is currently flooded with motorcycle companies, including Honda, Yamaha, Suzuki and Kawasaki, which are constantly competing to win consumer interest in buying motorcycle products. It is difficult for staff to set sales forecasts because of the volume of motorcycle sales data. In processing motorcycle sales data, it is necessary to carry out a data grouping technique. Using the K-Means method will divide and classify data into predetermined clusters by grouping based on certain classes. A data that has the closest similarity will be in the same cluster. Processing motorcycle sales data with this method can determine the amount of sales. The tools used in this research are Google Colaboratory. In this study there are 3 clusters. The results of this study are cluster 1 (many) there are 4 data in data 1, 2, 3 and 4, in cluster 2 (moderate) there are 3 data in data 5, 6 and 7, and cluster 3 (less) there are 5 data in data 8, 9, 10, 11 and 12. In the application of data mining using google colaboratory there are 3 colors namely yellow (cluster 1) there are 4 data, purple (cluster 2) there are 3 data and blue (cluster 3) there are 5 data. So it can be concluded that manual calculations using the k-means clustering method and the application of data mining using the Google colaboratory are relevant.

HUHgle: An Interactive Substrate Design Tool for Covalent Protein-ssDNA Labeling Using HUH-Tags.

HUH-tags have emerged as versatile fusion partners that mediate sequence specific protein-ssDNA bioconjugation through a simple and efficient reaction. Here we present HUHgle, a python-based interactive tool for the visualization, design, and optimization of substrates for HUH-tag mediated covalent labeling of proteins of interest with ssDNA substrates of interest. HUHgle streamlines design processes by integrating an intuitive plotting interface with a search function capable of predicting and displaying protein-ssDNA bioconjugate formation efficiency and specificity in proposed HUH-tag/ssDNA sequence combinations. Validation demonstrates that HUHgle accurately predicts product formation of HUH-tag mediated bioconjugation for single- and orthogonal-labeling reactions. In order to maximize the accessibility and utility of HUHgle, we have implemented it as a user-friendly Google Colab notebook which facilitates broad use of this tool, regardless of coding expertise.

Crop Type Classification using Sentinel 2A-Derived Normalized Difference Red Edge Index (NDRE) and Machine Learning Approach

Satellite remote sensing (RS) enables the extraction of vital information on land cover and crop type. Land cover and crop type classification using RS data and machine learning (ML) techniques have recently gained considerable attention in the scientific community. This study aimed to enhance remote sensing research using high-resolution satellite imagery and a ML approach. To achieve this objective, ML algorithms were employed to demonstrate whether it was possible to accurately classify various crop types within agricultural areas using the Sentinel 2A-derived Normalized Difference Red Edge Index (NDRE). Five ML classifiers, namely Support Vector Machines (SVM), Random Forest (RF), Decision Tree (DT), K-Nearest Neighbors (KNN), and Multi-Layer Perceptron (MLP), were implemented using Python programming on Google Colaboratory. The target land cover classes included cereals, fallow, forage, fruits, grassland-pasture, legumes, maize, sugar beet, onion-garlic, sunflower, and watermelon-melon. The classification models exhibited strong performance, evidenced by their robust overall accuracy (OA). The RF model outperformed, with an OA rate of 95% and a Kappa score of 92%. It was followed by DT (88%), KNN (87%), SVM (85%), and MLP (82%). These findings showed the possibility of achieving high classification accuracy using NDRE from a few Sentinel 2A images. This study demonstrated the potential enhancement of the application of high-resolution satellite RS data and ML for crop type classification in regions that have received less attention in previous studies.

Cyber-physical recreation of six DOF industrial robot arm

AbstractThe Cyber-Physical and Vehicle Manufacturing Laboratory, a model Industry 4.0 laboratory, is applying new innovative solutions to improve the quality of education. As part of this, a digital twin of the lab was designed and built, where users can practice. In the virtual space, it is possible to apply the known robot motion types, and the tool centre and wrist speed have been measured virtually. Robot control tasks can be performed “offline” using parameters. This information can then be transferred to the actual physical robot unit. The stable diffusion 1.5 deep learning model generates 2D geometric shapes for trajectory, allowing users to perform unique tasks during education. The Google Colab cloud-based service was used to teach our rendered-type dataset. For the 3D simulation frame, we used V-REP, which was developed on a desktop PC equipped with an Intel Core i5 7600K processor, Nvidia GTX1070 VGA with 8 GB of DDR5 VRAM, and 64 GB of DDR4 memory modules. The following material describes an existing industrial six-axis robot arm and its implementation, which can be controlled and programmed while performing virtual measurements after integrating into a Cyber-Physical system and using deep learning techniques.

Enhancing Brain Tumor Classification in MRI: Leveraging Deep Convolutional Neural Networks for Improved Accuracy

Brain tumors are among the deadliest forms of cancer, and there is a significant death rate in patients. Identifying and classifying brain tumors are critical steps in understanding their functioning. The best way to treat a brain tumor depends on its type, size, and location. In the modern era, Radiologists utilize Brain tumor locations that can be determined using magnetic resonance imaging (MRI). However, manual tests and MRI examinations are time-consuming and require skills. In addition, misdiagnosis of tumors can lead to inappropriate medical therapy, which could reduce their chances of living. As technology advances in Deep Learning (DL), Computer Assisted Diagnosis (CAD) as well as Machine Learning (ML) technique has been developed to aid in the detection of brain tumors, radiologists can now more accurately identify brain tumors. This paper proposes an MRI image classification using a VGG16 model to make a deep convolutional neural network (DCNN) architecture. The proposed model was evaluated with two sets of brain MRI data from Kaggle. Considering both datasets during the training at Google Colab, the proposed method achieved significant performance with a maximum overall accuracy of 96.67% and 97.67%, respectively. The proposed model was reported to have worked well during the training period and been highly accurate. The proposed model's performance criteria go beyond existing techniques.

Scanorama: integrating large and diverse single-cell transcriptomic datasets.

Merging diverse single-cell RNA sequencing (scRNA-seq) data from numerous experiments, laboratories and technologies can uncover important biological insights. Nonetheless, integrating scRNA-seq data encounters special challenges when the datasets are composed of diverse cell type compositions. Scanorama offers a robust solution for improving the quality and interpretation of heterogeneous scRNA-seq data by effectively merging information from diverse sources. Scanorama is designed to address the technical variation introduced by differences in sample preparation, sequencing depth and experimental batches that can confound the analysis of multiple scRNA-seq datasets. Here we provide a detailed protocol for using Scanorama within a Scanpy-based single-cell analysis workflow coupled with Google Colaboratory, a cloud-based free Jupyter notebook environment service. The protocol involves Scanorama integration, a process that typically spans 0.5-3 h. Scanorama integration requires a basic understanding of cellular biology, transcriptomic technologies and bioinformatics. Our protocol and new Scanorama-Colaboratory resource should make scRNA-seq integration more widely accessible to researchers.

Deteksi Email Spam dengan Continuous Bag-Of-Words dan Random Forest

Spam email poses a significant cyber threat, as scammers employ various tactics to deceive individuals into divulging sensitive information or downloading harmful content. For instance, in June 2023, Indonesia encountered approximately 6.51 thousand spam attacks, underscoring the widespread nature of this issue. These attacks frequently involve deceptive strategies, such as impersonation or false promises of rewards, to ensnare unsuspecting victims. Succumbing to spam can result in financial losses and other grave repercussions. To address this concern, this research addresses this pressing problem by focusing on email content classification to detect phishing attempts. The proposed solution leverages runtime platforms such as Google Colab and uses Continuous Bag of Words (CBOW) analysis and Random Forest methods. CBOW is selected for its effectiveness in capturing semantic relationships between words, allowing the model to extract meaningful features from the email content. Random Forest, on the other hand, is chosen for its ability to handle imbalanced datasets commonly encountered in email classification tasks, ensuring fair representation of both spam and ham emails during model training. By combining these two techniques, we aim to develop a robust classification model capable of accurately distinguishing between phishing (spam) and legitimate (ham) emails, thus enhancing email security measures. Through our approach, we aim to classify the SpamAssassin dataset into ham or spam categories, with an anticipated precision rate of 0.98, demonstrating the model's effectiveness in accurately identifying phishing emails.

Semantic Segmentation of Lungs using U-Net

Semantic segmentation of medical images is crucial for aiding radiologists and clinicians in accurately diagnosing conditions and planning treatments. This research introduces an innovative technique for the semantic segmentation of lungs using a U-Net design. The U-Net model demonstrates outstanding performance in various medical image segmentation tasks by effectively capturing both local and global features. Additionally, Res- U-Net architecture is harnessed to identify lung regions from lungs CT images, with a focus on improving the accuracy and efficiency of the segmentation process. Python is utilized for both designing and executing the algorithm in this work. Google Colab, a popular platform, is being used for its computational resources and collaboration features. The datasets were sourced from Iraq-Oncology Teaching Hospital with 70% of the data allocated for training and 30% for testing.

Brain Tumor Identification using YOLO Network

CAD systems for brain MRI analysis employ various AI techniques to assist radiologists in interpreting images and detecting abnormalities. These systems must be trained on large datasets encompassing diverse brain pathologies to ensure accurate detection and classification of different diseases. In this research, the use of YOLOv4 and YOLOv5 architectures for brain tumour detection in MRI images is an interesting application of deep learning technology. The performances metrices such as Precision, Recall, F1 Score and mAP are analysed. The coding for this work was developed using Python, utilizing TensorFlow as the platform. Simulations were carried out on Google Colab.

Classifications of CNV, DME, Drusen and Normal in Retinal Optical Coherence using CNN

In this research, the importance of Optical Coherence Tomography (OCT) in diagnosing and monitoring various retinal disorders, including Drusen, Diabetic Macular Edema (DME), and Choroidal Neovascularization (CNV), is highlighted. These conditions can have a significant impact on retinal health and vision. The research presents a technique that utilizes batch normalization for preprocessing OCT images. For classification of retinal disorders, the research employs the Inception v3 architecture, which is known for its effectiveness in image classification tasks. The performance of the proposed technique is evaluated using performance metrics such as sensitivity, specificity, accuracy, and precision. In this work, a total of 3,133 images were obtained from Kaggle.com. Among these, 710 images were classified as CNV, 895 as DME, 725 as drusen, and 804 as normal retinal images. Python was used for both designing and Google colab was used for executing the algorithm.

Comparing hybrid models for recognising objects in thermal images at nighttime

This research aims to revolutionize urban object recognition by developing cloud-based Python programs using intelligent algorithms. Unlike current models that focus on colour enhancement in nighttime thermal images, this work addresses the critical challenge of accurate object detection in urban landscapes. The proposed method incorporates a binary generative adversarial network (GAN) generator that can switch bidirectionally between daytime colour (DC) and nighttime infrared (NTIR) images. memory-based visual image memory (MVAM), system extracts important descriptive information from urban landscape images, reducing problems related to small sample sizes. This discussion presents a comprehensive improvement and evaluation of a deep learning image classification pipeline using Google Colab, demonstrating advanced image processing. Using TensorFlow, Keres and scikit image libraries combined with advanced algorithms such as DenseNet121 and MobileNetV2 presents a clear approach. We created a Bidirectional GAN + MVAM for object recognition in this work. Our method performed well, with an accuracy of 81.43%, precision of 51.16, recall of 50.11, and F-score of 46.37. The systematic presentation of the code presents a careful strategy to ensure optimal performance, stability, and efficiency of deep learning and image processing tasks.

Detection of Blood Cancer Cells using Microscopic Image

For the automatic diagnosis and classification of leukemia and leukemoid reactions, the IDB2 (acute lymphoblastic leukemia-image database) dataset has been utilized. This paper focuses on an automated method to differentiate between leukemoid and leukemia reactions using images of blood smear. MobileNetV3 is employed to classify and count WBC types from segmented images. The BCCD (Blood Cell Count Detection) dataset, which contains 364 images of blood smear and 349 single WBC type images, has been used in this work. The image segmentation algorithm incorporates Fuzzy C-means clustering, the snake algorithm, and fusion rules. For classification, the VGG19 CNN (Convolutional Neural Network) architecture-based deep learning technique is implemented. Python and Google colab is used for designing and executing the algorithm respectively.

Lung Tumor Classification Optimizer with Augment Input Images

This work focuses on the accurate classification of lung tumors in Computed Tomography (CT) images to determine tumor type and stage, essential for guiding treatment decisions. In this research, pretrained model VGG19 and Generative Adversarial Networks (GANs) models are used to classify the lung tumor in order to enhance the classification accuracy by increasing the diversity of the training dataset. The dataset comprises 1097 CT image samples, with 120 classified as benign, 561 as malignant, and 416 as normal cases. The coding for this work was developed using Python, with TensorFlow as the deep learning platform, and simulations were conducted on Google Colab. Performance metrics such as accuracy, sensitivity, specificity, and F1 score are evaluated to assess the effectiveness of the classification model. The datasets were obtained from the Iraq-Oncology Teaching Hospital.

Quantitative Assessment of Volcanic Thermal Activity from Space Using an Isolation Forest Machine Learning Algorithm

Understanding the dynamics of volcanic activity is crucial for volcano observatories in their efforts to forecast volcanic hazards. Satellite imager data hold promise in offering crucial insights into the thermal behavior of active volcanoes worldwide, facilitating the assessment of volcanic activity levels and identifying significant changes during periods of volcano unrest. The Moderate Resolution Imaging Spectroradiometer (MODIS) sensor, aboard NASA’s Terra and Aqua satellites, provides invaluable data with high temporal and spectral resolution, enabling comprehensive thermal monitoring of eruptive activity. The accuracy of volcanic activity characterization depends on the quality of models used to relate the relationship between volcanic phenomena and target variables such as temperature. Under these circumstances, machine learning (ML) techniques such as decision trees can be employed to develop reliable models without necessarily offering any particular or explicit insights. Here, we present a ML approach for quantifying volcanic thermal activity levels in near real time using thermal infrared satellite data. We develop an unsupervised Isolation Forest machine learning algorithm, fully implemented in Google Colab using Google Earth Engine (GEE) which utilizes MODIS Land Surface Temperature (LST) data to automatically retrieve information on the thermal state of volcanoes. We evaluate the algorithm on various volcanoes worldwide characterized by different levels of volcanic activity.

Pan-cancer G2C-Pro: A two-stage Gaussian clustering to prognostically stratify patients with advanced tumors treated with immune checkpoint inhibitors.

e13605 Background: Immune checkpoint inhibitors (ICIs) have revamped the clinical outcomes of patients (pts) affected by advanced solid tumors. The aim of our study is to develop a two-stage Gaussian Clustering algorithm (G2C) integrated with a logistic regression model (Pro) to prognostically stratifypts with advanced solid tumors treated with ICIs (pan-cancer G2C-Pro) based on baseline features. Methods: Data extraction involved, retrospectively, pts with advanced solid tumors (lung, melanoma, renal cancer, head and neck, urothelial carcinoma) treated with ICIs at Department of Medical Oncology, Ancona. Baseline body mass composition (BC) was assessed through computed tomography (CT) scan at L3 level, abstracting subcutaneous and visceral fat and muscle mass indicators. Moreover, baseline clinicopathologic features, nutritional status through The Controlling Nutritional Status (CONUT) score, and comorbidities were collected. An unsupervised clustering analysis was used to identify two groups of pts with different BC phenotype risk groups. Then, another unsupervised clustering analysis was used to identify two groups of pts within the dataset that hold prognostic significance according to BC risk groups, clinicopathologic features, CONUT score, and comorbidities. G2C was constructed by sequentially integrating two Gaussian Mixture models, each employing K-means++ initialization. Next, Pro was used to predict clusters’ label. The metrics used to evaluate the average performance of Pro on test sets were ACCuracy (ACC) and Area Under the Curve (AUC). The model was developed in Python on-cloud using the Google Colab service. Results: A total of 87 pts with complete data available were included in the final analysis. The two generated clusters for BC phenotype were BC_Low_Risk (n = 39) and BC_High_Risk (n = 48). Then, considering BC risk groups, clinicopathologic features, nutritional status, and comorbidities, two generated clusters were cluster 1 (n = 55) and cluster 2 (n = 32). Looking at clinical outcomes, median progression free survival was 16.1 months for cluster 1 versus 7.1 months for cluster 2 (HR: 0.57, 95% CI: 0.33-0.98, p = 0.04), and median overall survival was 41.8 months for cluster 1 versus 10.5 months for cluster 2 (HR: 0.47, 95% CI: 0.27-0.83, p = 0.008). The average ACC and AUC across all splits achieved by Pro model for patient classification into clusters were 0.94 and 0.89, respectively. From the feature ranking, it emerged that the one with higher importance was CONUT score, followed by BC phenotype risk groups and neutrophil-to-lymphocyte ratio. Conclusions: By using an easy-to-obtain and reproducible baseline BC, clinicopathologic, nutritional features, and comorbidities, we demonstrated that pan-cancer G2C-Pro is a promising and accurate prognostic model for stratifying pts with advanced tumors treated with ICIs.