Preliminary Evidence of the Use of Generative AI in Health Care Clinical Services: Systematic Narrative Review

JMIR Medical Informatics, Journal Year: 2024, Volume and Issue: 12, P. e52073 - e52073

Published: Jan. 30, 2024

Background Generative artificial intelligence tools and applications (GenAI) are being increasingly used in health care. Physicians, specialists, other providers have started primarily using GenAI as an aid or tool to gather knowledge, provide information, train, generate suggestive dialogue between physicians patients patients’ families friends. However, unless the use of is oriented be helpful clinical service encounters that can improve accuracy diagnosis, treatment, patient outcomes, expected potential will not achieved. As adoption continues, it essential validate effectiveness infusion intelligent technology understand gap actual GenAI. Objective This study synthesizes preliminary evidence on how assists, guides, automates rendering care The review scope was limited articles published peer-reviewed medical journals. Methods We screened selected 0.38% (161/42,459) January 1, 2020, May 31, 2023, identified from PubMed. followed protocols outlined PRISMA (Preferred Reporting Items for Systematic Reviews Meta-Analyses) guidelines select highly relevant studies with at least 1 element use, evaluation, validation services. were classified based their relevance functions activities descriptive analytical information presented articles. Results Of 161 articles, 141 (87.6%) reported assist services through knowledge access, collation, filtering. disease detection (19/161, 11.8%), diagnosis (14/161, 8.7%), screening processes (12/161, 7.5%) areas radiology (17/161, 10.6%), cardiology 7.5%), gastrointestinal medicine (4/161, 2.5%), diabetes (6/161, 3.7%). literature synthesis this suggests mainly diagnostic processes, improvement accuracy, purposes access. Although solves problem access may toward higher value creation Conclusions informs rather than assisting automating There service, but has yet actualized More service–level streamline some provides more automated help only retrieval needed. To transform purported, related must automate guide human-performed keep up optimism forward-thinking organizations take advantage

Language: Английский

---

Management of locally advanced non‐small cell lung cancer: State of the art and future directions

Cancer Communications, Journal Year: 2023, Volume and Issue: 44(1), P. 23 - 46

Published: Nov. 20, 2023

Abstract Lung cancer is the second most common and deadliest type of worldwide. Clinically, non‐small cell lung (NSCLC) pathological cancer; approximately one‐third affected patients have locally advanced NSCLC (LA‐NSCLC, stage III NSCLC) at diagnosis. Because its heterogeneity, LA‐NSCLC often requires multidisciplinary assessment. Moreover, prognosis much below satisfaction, efficacy traditional therapeutic strategies has reached a plateau. With emergence targeted therapies immunotherapies, as well continuous development novel radiotherapies, we entered an era treatment paradigm for LA‐NSCLC. Here, reviewed landscape relevant modalities, including adjuvant, neoadjuvant, perioperative immune in with resectable with/without oncogenic alterations; combinations chemoradiation immunotherapy/targeted therapy unresectable We addressed unresolved challenges that remain field, examined future directions to optimize clinical management increase cure rate

Language: Английский

---

A Lung CT Foundation Model Facilitating Disease Diagnosis and Medical Imaging

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 15, 2025

Abstract The concomitant development and evolution of lung computed tomography (CT) artificial intelligence (AI) has allowed non-invasive imaging to be a key part the clinical care patients with major diseases, such as cancer. However, paucity labeled CT data limited training highly efficacious AI models thereby retarded broad-scale adoption deployment AI-based in real-world setting. In this paper, We introduce LCTfound, foundational model that encodes images along correlated information, into neural network. LCTfound used self-supervised learning pre-trained by diffusion using large dataset containing 105,184 scans (totaling more than 28 million images) from multiple centers. was evaluated on 8 categories tasks, ranging scanning-level diagnosis pixel-level image restoration, including segmentation mediastinal neoplasm, pulmonary alveolar proteinosis, prognosis non-small cell cancer, prediction pathological response neoadjuvant chemoimmunotherapy, whole 3D modeling for surgical navigation, virtual angiography(CTA), reconstruction sparse views, enhancement low-dose images. Equipped robust few-shot capability, outperformed previously state-of-the-art all above tasks. is advancements representation CT, laying groundwork operates high efficacy across spectrum low-level high-level tasks serving dual purpose aiding diseases improving quality imaging.

Language: Английский

---

Neoadjuvant Immunotherapy: A Promising New Standard of Care

International Journal of Molecular Sciences, Journal Year: 2023, Volume and Issue: 24(14), P. 11849 - 11849

Published: July 24, 2023

Neoadjuvant immunotherapy has emerged as a promising approach in the treatment of various malignancies, with preclinical studies showing improved immune responses preoperative setting. FDA-approved neoadjuvant-immunotherapy-based approaches include triple-negative breast cancer and early non-small cell lung on basis improvement pathological response event free survival. Nevertheless, current trials have only shown benefits fraction patients. It is therefore crucial to identify predictive biomarkers improve patient selection for such approaches. This review aims provide an overview potential neoadjuvant cancer, bladder melanoma, colorectal gastric cancer. By extrapolation metastatic setting, we explore known biomarkers, i.e., PD-L1, mismatch repair deficiency tumour mutational burden, well early-disease-specific biomarkers. We also discuss challenges identifying reliable need standardized protocols guidelines their validation clinical implementation.

Language: Английский

---

Radiomics model based on intratumoral and peritumoral features for predicting major pathological response in non-small cell lung cancer receiving neoadjuvant immunochemotherapy

Frontiers in Oncology, Journal Year: 2024, Volume and Issue: 14

Published: March 20, 2024

Objective To establish a radiomics model based on intratumoral and peritumoral features extracted from pre-treatment CT to predict the major pathological response (MPR) in patients with non-small cell lung cancer (NSCLC) receiving neoadjuvant immunochemotherapy. Methods A total of 148 NSCLC who underwent immunochemotherapy two centers (SRRSH ZCH) were retrospectively included. The SRRSH dataset (n=105) was used as training internal validation cohort. Radiomics (T) regions (P1 = 0-5mm, P2 5-10mm, P3 10-15mm) CT. Intra- inter- class correlation coefficients least absolute shrinkage selection operator feature selection. Four single ROI models mentioned above combined (CR: T+P1+P2+P3) established by using machine learning algorithms. Clinical factors selected construct radiomics-clinical (CRC) model, which validated external center ZCH (n=43). performance assessed DeLong test, calibration curve decision analysis. Results Histopathological type only independent clinical risk factor. CR eight demonstrated good predictive (AUC=0.810) significantly improved than T (AUC=0.810 vs 0.619, p<0.05). CRC yielded best capability (AUC=0.814) obtained satisfactory test set (AUC=0.768, 95% CI: 0.62-0.91). Conclusion We that incorporates histopathological type, providing an effective approach for selecting suitable

Language: Английский

---

Non-invasive multimodal CT deep learning biomarker to predict pathological complete response of non-small cell lung cancer following neoadjuvant immunochemotherapy: a multicenter study

Journal for ImmunoTherapy of Cancer, Journal Year: 2024, Volume and Issue: 12(9), P. e009348 - e009348

Published: Sept. 1, 2024

Objectives Although neoadjuvant immunochemotherapy has been widely applied in non-small cell lung cancer (NSCLC), predicting treatment response remains a challenge. We used pretreatment multimodal CT to explore deep learning-based image biomarkers. Methods This study retrospectively obtained non-contrast enhanced and contrast enhancedbubu scans of patients with NSCLC who underwent surgery after receiving at multiple centers between August 2019 February 2023. Deep learning features were extracted from both construct the predictive models (LUNAI-uCT model LUNAI-eCT model), respectively. After feature fusion these two types features, fused (LUNAI-fCT model) was constructed. The performance evaluated using area under receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive value, negative value. SHapley Additive exPlanations analysis quantify impact imaging on prediction. To gain insights into how our makes predictions, we employed Gradient-weighted Class Activation Mapping generate saliency heatmaps. Results training validation datasets included 113 Center A 8:2 ratio, test dataset 112 (Center B n=73, C n=20, D n=19). In dataset, LUNAI-uCT, LUNAI-eCT, LUNAI-fCT achieved AUCs 0.762 (95% CI 0.654 0.791), 0.797 0.724 0.844), 0.866 0.821 0.883), Conclusions By extracting CT, constructed as an biomarker, which can non-invasively predict pathological complete for NSCLC.

Language: Английский

---

Immunotherapy for early-stage non-small cell lung cancer: A system review

Journal of Cancer Research and Therapeutics, Journal Year: 2023, Volume and Issue: 19(4), P. 849 - 865

Published: Aug. 1, 2023

ABSTRACT With the addition of immunotherapy, lung cancer, one most common cancers with high mortality rates, has broadened treatment landscape. Immune checkpoint inhibitors have demonstrated significant efficacy in non-small cell cancer (NSCLC) and are now used as first-line therapy for metastatic disease, consolidation after radiotherapy unresectable locally advanced adjuvant surgical resection chemotherapy resectable disease. The use neoadjuvant immunotherapy patients early-stage NSCLC, however, is still debatable. We will address several aspects, namely initial monotherapy, combination chemotherapy, immunotherapy-related biomarkers, adverse effects, ongoing randomized controlled trials, current issues future directions NSCLC be discussed here.

Language: Английский

---

A CT-based deep learning model predicts overall survival in patients with muscle invasive bladder cancer after radical cystectomy: a multicenter retrospective cohort study

International Journal of Surgery, Journal Year: 2024, Volume and Issue: unknown

Published: Feb. 13, 2024

Background: Muscle invasive bladder cancer (MIBC) has a poor prognosis even after radical cystectomy (RC). Postoperative survival stratification based on radiomics and deep learning algorithms may be useful for treatment decision-making follow-up management. This study was aimed to develop validate (DL) model preoperative CT predicting post-cystectomy overall in patients with MIBC. Methods: MIBC who underwent RC were retrospectively included from four centers, divided into the training, internal validation external sets. A incorporated convolutional block attention module (CBAM) built using images. We assessed prognostic accuracy of DL compared it classic handcrafted clinical model. Then, nomogram (DLRN) developed by combining clinicopathological factors, score (Rad-score) (DL-score). Model performance C-index, KM curve, time-dependent ROC curve. Results: total 405 this study. The DL-score achieved much higher C-index than Rad-score (0.690 vs. 0.652 0.618 set, 0.658 0.601 0.610 set). After adjusting clinicopathologic variables, identified as significantly independent risk factor OS multivariate Cox regression analysis all sets (all P <0.01). DLRN further improved performance, 0.713 (95%CI: 0.627-0.798) set 0.685 0.586-0.765) respectively. Conclusions: can predict outcome MIBC, which help guide

Language: Английский

---

CT-based quantification of intratumoral heterogeneity for predicting pathologic complete response to neoadjuvant immunochemotherapy in non-small cell lung cancer

Frontiers in Immunology, Journal Year: 2024, Volume and Issue: 15

Published: June 12, 2024

To investigate the prediction of pathologic complete response (pCR) in patients with non-small cell lung cancer (NSCLC) undergoing neoadjuvant immunochemotherapy (NAIC) using quantification intratumoral heterogeneity from pre-treatment CT image.

Language: Английский

---

Combined deep learning and radiomics in pretreatment radiation esophagitis prediction for patients with esophageal cancer underwent volumetric modulated arc therapy

Radiotherapy and Oncology, Journal Year: 2024, Volume and Issue: 199, P. 110438 - 110438

Published: July 14, 2024

Language: Английский

---