Hasil untuk "Neoplasms. Tumors. Oncology. Including cancer and carcinogens"

Eric R. Lutz, Annie A Wu, Elaine O Bigelow et al.

Chen S, Gao PJ

Shuang Chen,1,2 Peng-ji Gao1,2 1Department of General Surgery, Beijing Jishuitan Hospital, Capital Medical University, Beijing, 100035, People’s Republic of China; 2Beijing Research Institute of Traumatology and Orthopaedics, Beijing Jishuitan Hospital, Capital Medical University, Beijing, 100035, People’s Republic of ChinaCorrespondence: Peng-ji Gao, Department of General Surgery, Beijing Jishuitan Hospital, Capital Medical University, No. 31, Xinjiekou East Street, Xicheng District, Beijing, 100035, People’s Republic of China, Tel +8601058398275, Email gaopengji@mail.ccmu.edu.cnBackground: Colorectal cancer (CRC) is one of the most common and deadly malignancies worldwide. Sarcopenia, defined as a progressive loss of skeletal muscle mass and function, has recently been recognized as an important prognostic factor in CRC, influencing both postoperative complications and long-term survival.Methods: We conducted a descriptive review of 18 clinical studies investigating the association between sarcopenia and CRC across stages I–IV. Sarcopenia was primarily assessed using computed tomography-derived skeletal muscle index (SMI) or psoas index (PI) at the lumbar vertebrae (L3/L4), with some studies additionally incorporating muscle strength and performance.Results: The prevalence of sarcopenia among CRC patients ranged from 12% to 60%. Most studies reported higher risks of postoperative complications in sarcopenic patients. For instance, Peng et al demonstrated an increased risk of complications in stage IV CRC patients with sarcopenia (OR: 3.12, 95% CI: 1.14– 8.49). Regarding survival, sarcopenia was consistently associated with worse overall survival (OS), disease-free survival (DFS), and recurrence-free survival (RFS). Brown et al (2018) showed that deterioration in muscle mass and radiodensity significantly predicted poorer OS in 1,924 stage I–III CRC patients (HR: 2.15, 95% CI: 1.59– 2.92). However, several studies reported no significant associations.Conclusion: Sarcopenia is prevalent in CRC patients and strongly correlates with both short-term surgical outcomes and long-term prognosis. However, current evidence is mainly derived from heterogeneous observational studies, and further prospective studies are needed before sarcopenia assessment can be translated into routine clinical practice.Keywords: colorectal cancer, sarcopenia, skeletal muscle index, prognosis, postoperative complications, survival

C. Hotnog, M. Bostan, Matei Anghelescu et al.

Head and neck squamous cell carcinoma (HNSCC) is an aggressive cancer with a complex molecular landscape. Despite extensive research, our understanding of the molecular mechanisms remains incomplete, hindering the development of effective therapeutic strategies for this disease. Long non-coding RNAs (lncRNAs) have emerged as crucial factors in cancer biology, regulating key networks across various malignancies. These molecules exert their regulatory functions through interactions with nucleic acids or proteins, thereby influencing signaling pathways within tumor cells. Consequently, lncRNAs play a significant role in key processes like cell proliferation, metastasis, immune evasion, and treatment resistance. This review offers a comprehensive overview of current knowledge regarding lncRNA-mediated mechanisms in HNSCC. The first section explores how lncRNAs influence tumor processes through various modulation mechanisms, including transcriptional and post-transcriptional regulation, chromatin remodeling, and epigenetic modifications. We also highlight the impact of lncRNAs on specific signaling pathways that control essential cellular functions (e.g., proliferation, apoptosis, angiogenesis, invasion, metastasis). Ultimately, this underscores the promising potential of lncRNAs as diagnostic biomarkers and therapeutic targets capable of enhancing patient care in oncology. Gaining a deep understanding of how lncRNAs modulate carcinogenic mechanisms may yield innovative approaches for early detection, personalized treatment, and improved clinical outcomes for HNSCC patients.

Fangmin Zhong, Fangyi Yao, Zihao Wang et al.

Abstract Acute myeloid leukemia (AML) remains challenging due to molecular heterogeneity and limited prognostic models integrating tumor microenvironment dynamics. While thrombocytopenia correlates with poor outcomes, the roles of platelet-related genes (PRGs) in AML pathogenesis are unclear. We integrated multiomics data to analyze platelet-clinicopathological associations, identify PRGs via weighted gene coexpression network analysis, and define molecular subtypes through unsupervised clustering. A machine learning-derived PRGScore model was developed and validated. Immune features were assessed using CIBERSORT, and drug responses via ex vivo profiling. The results showed that low platelet counts predicted a poor prognosis and were inversely correlated with blast proportions, suggesting suppression of leukemia-mediated megakaryopoiesis. Platelet recovery is linked to inflammatory and coagulation pathways. We identified 22 key PRGs that were downregulated in AML and enriched in immunomodulatory pathways. Unsupervised clustering stratified AML into three PRG-based subtypes: C1 (low PRGs/platelets, best survival), C2 (intermediate), and C3 (high PRGs/platelets, worst survival). C1 exhibited cytotoxic T-cell enrichment, reduced immune checkpoint expression, a high proportion of blasts and increased proliferative activity. C3 was characterized by myeloid immunosuppression, elevated checkpoint levels, and the activation of chemoresistance pathways. The PRGScore model demonstrated robust prognostic accuracy across the 10 cohorts. Patients with a high PRGScore had a significantly worse prognosis and sensitivity to immune checkpoint inhibitors, whereas patients with a low PRGScore had a better response to cytarabine and venetoclax. This study establishes PRGs as key regulators of AML biology and prognosis, bridging platelet dynamics, immune interactions, and machine learning. The PRGScore framework advances precision medicine through risk stratification, therapeutic targeting, and biomarker discovery.

Raquel Ciérvide, Roberto Manchón, Daniela Angel et al.

BACKGROUND: This study evaluates practices and preferences in treating intact brain metastases with stereotactic-radiosurgery (SRS) among members of the SEOR-SRS_SBRT working group, focusing on clinical protocols, equipment usage, and treatment parameters. MATERIALS AND METHODS: A survey conducted via Google Forms targeted 149 group members, with responses collected from one representative per institution between April and May 2024. Respondents included radiation oncologists from Mexico, Argentina, Portugal, and Spain, and data analysis covered demographics, equipment, treatment protocols, immobilization techniques, dose schedules, image-guided radiation therapy (IGRT), and prescription criteria. RESULTS: Out of 149 members, 28 institutions responded. Most participants (64.5%) had over 10 years of experience. Single-fraction-SRS was practiced by 64.5%, while fractionated SRS-SRT was used by 96.8%. Linear accelerators (C-Linac) were the primary equipment (86.7%). Specific protocols for brain metastases were reported by 80%. SRS was preferred for 1–3 metastases (93.3%), while whole-brain radiation therapy (WBRT) was used for > 10 metastases (70%). Considering the type of stereotactic localization, frameless systems were employed in 69% while rigid-frames were used in 31% of cases. The most common immobilization technique was a reinforced mask (50%). Planning computed tomography (CT)/magnetic resonance imaging (MRI) slice thickness ≤ 2 mm was standard, and automatic registration was applied in 69%. Doses of 21–23 Gy were common for lesions < 1 cm, while < 20 Gy was typical for 2–3 cm lesions. Margins for single-fraction SRS were 1 mm in 50% of cases. IGRT verification used cone-beam CT (64.5%) and surface-guided radiation therapy (35.5%). CONCLUSION: The findings reveal variability in SRS practice, particularly in immobilization, dose prescriptions, and IGRT methods, emphasizing the need for standardized guidelines to optimize patient outcomes and adapt treatments to institutional resources and patient-specific factors.

Xiafeng Wang, Zhangping Huang, Lin Peng et al.

Jun Liu, Jun Liu, Wenjun Zhang et al.

ObjectiveTo investigate the clinical characteristics and therapeutic approaches for granular cell tumors (GCT) of the neurohypophysis.Materials and methodsRetrospective case series and analyzed the clinical data of six patients with histopathologically confirmed GCT of the neurohypophysis, also conducting a simple review of relevant literature.ResultsThe median age at diagnosis for the cohort of six patients was 41.0 ± 11.73 years, with an age range of 21.8 to 52.7 years. A predominance of female patients was noted, accounting for five out of six cases. The most common clinical symptoms were headache and visual disturbances, each reported in five of the six patients. Magnetic resonance imaging (MRI) of the brain predominantly revealed a rounded morphology, and well-defined boundaries. Of these tumors, two were located in the suprasellar region while four were situated within the sellar region, encompassing intrasellar, suprasellar, and parasellar locations. Contrast-enhanced MRI demonstrated heterogeneous enhancement in four cases and homogeneous enhancement in two cases. Surgical intervention, either through a neuro-endoscopic endonasal transsphenoidal approach or craniotomy, achieved total or subtotal tumor resection in all patients. Postoperative histopathological examination confirmed the diagnosis of GCT in each instance. All patients participated in follow-up evaluations, during which varying degrees of clinical symptom improvement were documented. Importantly, none of the four patients who underwent complete tumor resection exhibited recurrence or metastasis.ConclusionGCT of the neurohypophysis are rarely encountered clinical practice. The definitive diagnosis of GCT primarily relies on histopathological evaluation. Currently, the standard therapeutic approach involves complete surgical excision of the tumor using a neuroendoscopic endonasal transsphenoidal technique. Post-resection, the rates of recurrence and metastasis are significantly low.

Eva Molnárová, Ties A. Mulders, Marcela Spee-Dropková et al.

Evolutionary therapy (ET) aims to steer tumor evolution by adjusting treatment timing and dosing to control rather than eradicate tumor burden. Clinical use requires reliable monitoring of tumor dynamics to inform mathematical models that guide therapy. In cancers such as metastatic castrate-resistant prostate cancer and relapsed platinum-sensitive ovarian cancer, ET models are informed by serial serum biomarkers. For cancers lacking reliable biomarkers, such as metastatic non-small cell lung cancer (NSCLC), radiographic imaging remains the primary method for treatment response assessment, typically using RECIST 1.1 criteria. RECIST, which tracks a few lesions with one-dimensional (1D) measurements and defines progression relative to the nadir, the smallest tumor burden recorded after treatment, was not designed to support ET. It may miss early regrowth, underrepresent tumor burden, and obscure disease trends. Using a virtual NSCLC patient model, we demonstrate that lesion selection and measurement dimensionality strongly affect progression detection. Two-dimensional metrics provide modest improvement, but only 3D volumetric measurements accurately capture both tumor burden and its dynamics, which are key requirements for ET. To support ET in cancers lacking biomarkers, response assessment must evolve beyond RECIST by integrating volumetric imaging, automated segmentation, and potentially liquid biopsies, alongside redefining progression criteria to enable adaptive, patient-centered treatments.

Rupesh Dulal, Rabin Dulal

Identifying the extent of brain tumors is a significant challenge in brain cancer treatment. The main difficulty is in the approximate detection of tumor size. Magnetic resonance imaging (MRI) has become a critical diagnostic tool. However, manually detecting the boundaries of brain tumors from MRI scans is a labor-intensive task that requires extensive expertise. Deep learning and computer-aided detection techniques have led to notable advances in machine learning for this purpose. In this paper, we propose a modified You Only Look Once (YOLOv8) model to accurately detect the tumors within the MRI images. The proposed model replaced the Non-Maximum Suppression (NMS) algorithm with a Real-Time Detection Transformer (RT- DETR) in the detection head. NMS filters out redundant or overlapping bounding boxes in the detected tumors, but they are hand-designed and pre-set. RT-DETR removes hand-designed components. The second improvement was made by replacing the normal convolution block with ghost convolution. Ghost Convolution reduces computational and memory costs while maintaining high accuracy and enabling faster inference, making it ideal for resource-constrained environments and real-time applications. The third improvement was made by introducing a vision transformer block in the backbone of YOLOv8 to extract context-aware features. We used a publicly available dataset of brain tumors in the proposed model. The proposed model performed better than the original YOLOv8 model and also performed better than other object detectors (Faster R- CNN, Mask R-CNN, YOLO, YOLOv3, YOLOv4, YOLOv5, SSD, RetinaNet, EfficientDet, and DETR). The proposed model achieved 0.91 mAP (mean Average Precision)@0.5.

Majid Farhadloo, Arun Sharma, Alexey Leontovich et al.

Given multi-type point maps from different place-types (e.g., tumor regions), our objective is to develop a classifier trained on the source place-type to accurately distinguish between two classes of the target place-type based on their point arrangements. This problem is societally important for many applications, such as generating clinical hypotheses for designing new immunotherapies for cancer treatment. The challenge lies in the spatial variability, the inherent heterogeneity and variation observed in spatial properties or arrangements across different locations (i.e., place-types). Previous techniques focus on self-supervised tasks to learn domain-invariant features and mitigate domain differences; however, they often neglect the underlying spatial arrangements among data points, leading to significant discrepancies across different place-types. We explore a novel multi-task self-learning framework that targets spatial arrangements, such as spatial mix-up masking and spatial contrastive predictive coding, for spatially-delineated domain-adapted AI classification. Experimental results on real-world datasets (e.g., oncology data) show that the proposed framework provides higher prediction accuracy than baseline methods.

Yakup Ergun

Abstract Introduction Metastasis represents the leading cause of cancer-related mortality and is characterized by complex biological processes such as genomic instability, immune evasion, and therapy resistance. While metastatic tumors often retain the truncal drivers of their primary counterparts, the extent and nature of additional somatic alterations acquired during progression remain incompletely defined across cancer types. Methods A comprehensive pan-cancer analysis was conducted using targeted sequencing data from 2,846 patients with matched primary and metastatic tumors (totaling 5,692 samples) obtained from the AACR Project GENIE v18.0 cohort. Harmonized variant calls were used to compare mutation burden, fraction of genome altered (FGA), gene-level mutation frequencies, copy number alterations (CNA), and structural variants (SV) between compartments. Statistical comparisons were adjusted for multiple testing using the Benjamini-Hochberg method. Results Metastatic tumors exhibited a significantly higher median mutation count (6 vs. 5; p < 0.001) and FGA (0.186 vs. 0.140; p < 0.001) compared to matched primary tumors. This increase was most prominent in non-small cell lung, breast, colorectal, pancreatic, and prostate cancers. Eleven genes, including KDM5A, CDKN2A, MYC, ESR1, and AR, were significantly enriched in metastases, suggesting mechanisms such as cell cycle deregulation, therapy-induced selection, and chromatin remodeling. Notably, ESR1 alterations were enriched in breast cancer metastases, consistent with endocrine therapy resistance, while AR alterations were markedly more frequent in metastatic prostate cancer. CNA analysis revealed recurrent amplifications (MYC, ERBB2, CCND1) and deletions (CDKN2A, PTEN, RB1) in metastatic tumors. Structural variants involving genes linked to DNA damage response and epigenetic regulation were also more prevalent in the metastatic setting. Conclusions In this large-scale matched cohort of 2,846 patients and 5,692 tumor samples, metastatic tumors exhibited increased mutation burden and widespread genomic instability. Although treatment data were not available to directly associate resistance-related alterations with specific therapies, the observed patterns suggest that these acquired changes reflect context-dependent selection for survival and proliferative advantage in advanced disease, rather than the emergence of novel metastasis-specific driver events.

E. Sanarova, A. Lantsova, L. Nikolaeva et al.

Introduction. Improving the efficacy of chemotherapy is a non-trivial task of modern oncology. Its successful solution requires knowledge in many fields, including physiology, pathology, clinical oncology, pharmacology and others. The search for small molecules that selectively kill tumor cells led to the accidental discovery of erastin.Text. Erastin is a unique molecule that has a quinazoline fragment in its structure. Not so long ago it became known that the antitumour effect of this compound is due to the induction of ferroptosis – an iron-dependent form of cell death caused by lipid peroxidation. Erastin is able to induce ferroptosis through various biochemical pathways, including blocking of cystine-glutamate transport channel of cell membrane and potential-dependent anion channel of mitochondria, as well as activation of p53 protein.Conclusion. Pharmacological induction of ferroptosis by erastin and its analogues represents a promising direction in cancer chemotherapy. In addition, erastin and its analogues are able to increase sensitivity to chemotherapy and radiation therapy, which allows us to talk about the possibility of their use in the combined treatment of malignant neoplasms.

Guang Li PhD, Victoria Yu PhD, Kaitlyn Ryan BS et al.

Purpose To improve the setup reproducibility of neck curvature using real-time optical surface imaging (OSI) guidance on 2 regions of interest (ROIs) to infer cervical spine (c-spine) curvature for surface-guided radiotherapy (SGRT) of head-and-neck (HN) and c-spine cancer. Methods A novel SGRT setup approach was designed to reproduce neck curvature with 2 ROIs: upper-chest ROI and open-face ROI. It was hypothesized that the neck curvature could be reproduced if both ROIs were aligned within ±3 mm/2˚ tolerance. This was tested prospectively in 7 volunteers using real-time 3D-OSI guidance and lateral 2D-photography verification after the 3D and 2D references were captured from the initial conventional setup. Real-time SGRT was performed to align chest-ROI and face-ROI, and the longitudinal distance between them was adjustable using a head-support slider. Verification of neck curvature anteriorly and posteriorly was achieved by overlaying edge-extracted lateral pictures. Retrospectively, the relationship between anterior surface and spinal canal alignment was checked in 11 patients using their simulation CT (simCT) and setup cone-beam CT (CBCT). After the anterior surface was rigidly aligned, the spinal canal alignment was checked and quantified using the mean-distance-to-agreement (MDA) and DICE similarity index, and surface-to-spine correlation was calculated. Results The reproducibility of neck curvatures using the 2xROI SGRT setup is verified and the mean neck-outline-matching difference is within ±2 mm in lateral photographic overlays. The chest-ROI alignment takes 110 ± 58 s and the face-ROI takes 60 ± 35 s. When the anterior body surface is aligned (MDA = 1.1 ± 0.6 mm, DICE = 0.96 ± 0.02,) the internal spinal canal is also aligned (MDA = 1.0 ± 0.3 mm, DICE = 0.84 ± 0.04) in 11 patients. The surface-to-spine correlation is c = 0.90 (MDA) and c = 0.85 (DICE). Conclusion This study demonstrates the feasibility of the novel 2-ROI SGRT setup technique to achieve reproducible neck and c-spine curvature regardless of neck visibility and availability as ROI. Staff training is needed to adopt this unconventional SGRT technique to improve patient setup.

B. Blanco, R. Palma, M. Hurtado et al.

Targeted therapeutic interventions utilizing low-inten\-sity ultrasound (LIUS) exhibit substantial potential for hindering the proliferation of cancer stem cells. This investigation introduces a multiscale model and computational framework to comprehensively explore the therapeutic LIUS on poroelastic tumor dynamics, thereby unraveling the intricacies of mechanotransduction mechanisms at play. Our model includes both macroscopic timescales encompassing days and rapid timescales spanning from microseconds to seconds, facilitating an in-depth comprehension of tumor behavior. We unveil the discerning suppression or reorientation of cancer cell proliferation and migration, enhancing a notable redistribution of cellular phases and stresses within the tumor microenvironment. Our findings defy existing paradigms by elucidating the impact of LIUS on cancer stem cell behavior. This endeavor advances our fundamental understanding of mechanotransduction phenomena in the context of LIUS therapy, thus underscoring its promising as a targeted therapeutic modality for cancer treatment. Furthermore, our results make a substantial contribution to the broader scientific community by shedding light on the intricate interplay between mechanical forces, cellular responses, and the spatiotemporal evolution of tumors. These insights hold the promising to promote a new perspective for the future development of pioneering and highly efficacious therapeutic strategies for combating cancer in a personalized manner.

Xinran Li, Yi Shuai, Chen Liu et al.

Tumor synthesis can generate examples that AI often misses or over-detects, improving AI performance by training on these challenging cases. However, existing synthesis methods, which are typically unconditional -- generating images from random variables -- or conditioned only by tumor shapes, lack controllability over specific tumor characteristics such as texture, heterogeneity, boundaries, and pathology type. As a result, the generated tumors may be overly similar or duplicates of existing training data, failing to effectively address AI's weaknesses. We propose a new text-driven tumor synthesis approach, termed TextoMorph, that provides textual control over tumor characteristics. This is particularly beneficial for examples that confuse the AI the most, such as early tumor detection (increasing Sensitivity by +8.5%), tumor segmentation for precise radiotherapy (increasing DSC by +6.3%), and classification between benign and malignant tumors (improving Sensitivity by +8.2%). By incorporating text mined from radiology reports into the synthesis process, we increase the variability and controllability of the synthetic tumors to target AI's failure cases more precisely. Moreover, TextoMorph uses contrastive learning across different texts and CT scans, significantly reducing dependence on scarce image-report pairs (only 141 pairs used in this study) by leveraging a large corpus of 34,035 radiology reports. Finally, we have developed rigorous tests to evaluate synthetic tumors, including Text-Driven Visual Turing Test and Radiomics Pattern Analysis, showing that our synthetic tumors is realistic and diverse in texture, heterogeneity, boundaries, and pathology.

Erik Blom, Stefan Engblom, Gesina Menz

We present a stochastic computational model of avascular tumors, emphasizing the detailed implementation of the first four so-called hallmarks of cancer: self-sufficiency in growth factors, resistance to growth inhibitors, avoidance of apoptosis, and unlimited growth potential. Our goal is to provide a foundational understanding of the first steps of cancer malignancy while addressing modeling uncertainties, thus bringing us closer to a first-principles grasp of this process. Preliminary numerical simulations illustrate the comprehensiveness of our perspective.

Fei Liu, Qing Xu, Parbatraj Regmi et al.

BackgroundLymphoepithelioma-like intrahepatic cholangiocarcinoma (LEL-ICC) is a rare variant of intrahepatic cholangiocarcinoma. Epstein–Barr virus (EBV) infection was considered to play a pivotal role in the tumorigenesis of LEL-ICC. It is difficult to diagnosis of LEL-ICC due to the lack of specific features regarding the laboratory test results and imaging findings. At present, the diagnosis of LEL-ICC mainly depends on the histopathologic and immunohistochemical examinations. In addition, the prognosis of LEL-ICC was better than classical cholangiocarcinomas. To our knowledge, only few cases of LEL-ICC have been reported in the literature.Case presentationWe presented a case of a 32-year-old Chinese female with LEL-ICC. She had a 6-month history of upper abdominal pain. The magnetic resonance imaging (MRI) showed a 1.1× 1.3 cm lesion in the left lobe of liver, appearing low signal intensity on T1-weighted images and high signal intensity on T2-weighted images. The patient underwent laparoscopic left lateral sectionectomy. The postoperative histopathologic and immunohistochemical examinations results allowed for the definitive diagnosis of LEL-ICC. The patient was free from tumor recurrence after a 28 months follow-up.ConclusionIn this study, we reported a rare case of LEL-ICC associated with both HBV and EBV infection. EBV infection might play a pivotal role in the carcinogenesis of LEL-ICC, and surgical resection is still the most effective treatment at present. Further research on the etiology and treatment strategies of LEL-ICC is required.

Zhengliang Liu, Peilong Wang, Yiwei Li et al.

This paper presents RadOnc-GPT, a large language model specialized for radiation oncology through advanced tuning methods. RadOnc-GPT was finetuned on a large dataset of radiation oncology patient records from the Mayo Clinic in Arizona. The model employs instruction tuning on three key tasks - generating radiotherapy treatment regimens, determining optimal radiation modalities, and providing diagnostic descriptions/ICD codes based on patient diagnostic details. Evaluations conducted by comparing RadOnc-GPT outputs to general large language model outputs showed higher ROUGE scores in these three tasks. The study demonstrated the potential of using large language models fine-tuned using domain-specific knowledge like RadOnc-GPT to achieve transformational capabilities in highly specialized healthcare fields such as radiation oncology. However, our model's clinical relevance requires confirmation, and it specializes in only the aforementioned three specific tasks and lacks broader applicability. Furthermore, its evaluation through ROUGE scores might not reflect the true semantic and clinical accuracy - challenges we intend to address in future research.

Jia Zhai, Hui Liu

High-throughput screening technology has facilitated the generation of large-scale drug responses across hundreds of cancer cell lines. However, there exists significant discrepancy between in vitro cell lines and actual tumors in vivo in terms of their response to drug treatments, because of tumors comprise of complex cellular compositions and histopathology structure, known as tumor microenvironment (TME), which greatly influences the drug cytotoxicity against tumor cells. To date, no study has focused on modeling the impact of the TME on clinical drug response. This paper proposed a domain adaptation network for feature disentanglement to separate representations of cancer cells and TME of a tumor in patients. Two denoising autoencoders were separately used to extract features from cell lines (source domain) and tumors (target domain) for partial domain alignment and feature decoupling. The specific encoder was enforced to extract information only about TME. Moreover, to ensure generalizability to novel drugs, we applied a graph attention network to learn the latent representation of drugs, allowing us to linearly model the drug perturbation on cellular state in latent space. We calibrated our model on a benchmark dataset and demonstrated its superior performance in predicting clinical drug response and dissecting the influence of the TME on drug efficacy.

Gayathri Girish, Ponnathota Spandana, Badrish Vasu

Objective: This paper proposes a deep learning model for breast cancer detection from reconstructed images of microwave imaging scan data and aims to improve the accuracy and efficiency of breast tumor detection, which could have a significant impact on breast cancer diagnosis and treatment. Methods: Our framework consists of different convolutional neural network (CNN) architectures for feature extraction and a region-based CNN for tumor detection. We use 7 different architectures: DenseNet201, ResNet50, InceptionV3, InceptionResNetV3, MobileNetV2, NASNetMobile and NASNetLarge and compare its performance to find the best architecture out of the seven. An experimental dataset of MRI-derived breast phantoms was used. Results: NASNetLarge is the best architecture which can be used for the CNN model with accuracy of 88.41% and loss of 27.82%. Given that the model's AUC is 0.786, it can be concluded that it is suitable for use in its present form, while it could be improved upon and trained on other datasets that are comparable. Impact: One of the main causes of death in women is breast cancer, and early identification is essential for enhancing the results for patients. Due to its non-invasiveness and capacity to produce high-resolution images, microwave imaging is a potential tool for breast cancer screening. The complexity of tumors makes it difficult to adequately detect them in microwave images. The results of this research show that deep learning has a lot of potential for breast cancer detection in microwave images