Application of Artificial Intelligence in the Diagnosis, Treatment, and Prognostic Evaluation of Mediastinal Malignant Tumors

Pang, Jiyun; Xiu, Weigang; Ma, Xuelei

doi:10.3390/jcm12082818

Cited by 10 publications

(5 citation statements)

References 123 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In its testing phase, the model exhibited an accuracy close to that of human clinicians in classifying images of lesions as malignant or non-malignant (Esteva et al, 2017). Later research would discover that the algorithm underlying this recognition system was relying less on visual characteristics of the lesion itself and more often on the presence or absence of a ruler in the sample image; malignant lesions in the training data typically were photographed near a ruler and benign ones were not (Pang, Xiu, & Ma, 2023). This obvious shortcoming was not evident in initial research on the system because the machine learning models cannot specify how they make categorization decisions.…”

Section: Assumption: Data In Need Of Coding or Classifyingmentioning

confidence: 90%

A special delivery by a fork: Where does artificial intelligence come from?

Thornton

2023

New Drctns Evaluation

View full text Add to dashboard Cite

In this article, I discuss the use of artificial intelligence (AI) in evaluation and its relevance to the evolution of the field. I begin with a background on how AI models are developed, including how machine learning makes sense of data and how the algorithms it develops go on to power AI models. I go on to explain how this foundational understanding of machine learning and natural language processing informs where AI might and might not be effectively used. A critical concern is that AI models are only as strong as the data on which they are trained, and evaluators should consider important limitations when using AI, including its relevance to structural inequality. In considering the relationship between AI and evaluation, evaluators must consider both AI's use as an evaluative tool and its role as a new subject of evaluation. As AI becomes more and more relevant to a wider array of fields and disciplines, evaluators will need to develop strategies for how good the AI is (or is not), and what good the AI might (or might not) do.

show abstract

Section: Assumption: Data In Need Of Coding or Classifyingmentioning

confidence: 90%

A special delivery by a fork: Where does artificial intelligence come from?

Thornton

2023

New Drctns Evaluation

View full text Add to dashboard Cite

show abstract

“…Radiomics has a great potential as a non-invasive biomarker to quantify several tumor characteristics, both standalone and combined with artificial intelligence methods such as machine learning [ 31 – 33 ]. However, it faces challenges to clinical implementation [ 34 ].…”

Section: Discussionmentioning

confidence: 99%

Effects of Interobserver Segmentation Variability and Intensity Discretization on MRI-Based Radiomic Feature Reproducibility of Lipoma and Atypical Lipomatous Tumor

Gitto,

Cuocolo,

Giannetta

et al. 2024

J Digit Imaging. Inform. med.

View full text Add to dashboard Cite

Segmentation and image intensity discretization impact on radiomics workflow. The aim of this study is to investigate the influence of interobserver segmentation variability and intensity discretization methods on the reproducibility of MRI-based radiomic features in lipoma and atypical lipomatous tumor (ALT). Thirty patients with lipoma or ALT were retrospectively included. Three readers independently performed manual contour-focused segmentation on T1-weighted and T2-weighted sequences, including the whole tumor volume. Additionally, a marginal erosion was applied to segmentations to evaluate its influence on feature reproducibility. After image pre-processing, with included intensity discretization employing both fixed bin number and width approaches, 1106 radiomic features were extracted from each sequence. Intraclass correlation coefficient (ICC) 95% confidence interval lower bound ≥ 0.75 defined feature stability. In contour-focused vs. margin shrinkage segmentation, the rates of stable features extracted from T1-weighted and T2-weighted images ranged from 92.68 to 95.21% vs. 90.69 to 95.66% after fixed bin number discretization and from 95.75 to 97.65% vs. 95.39 to 96.47% after fixed bin width discretization, respectively, with no difference between the two segmentation approaches (p ≥ 0.175). Higher stable feature rates and higher feature ICC values were found when implementing discretization with fixed bin width compared to fixed bin number, regardless of the segmentation approach (p < 0.001). In conclusion, MRI radiomic features of lipoma and ALT are reproducible regardless of the segmentation approach and intensity discretization method, although a certain degree of interobserver variability highlights the need for a preliminary reliability analysis in future studies.

show abstract

“…The lack of randomized studies has not allowed for drawing any strong conclusion on the real prognostic impact of metastatic thymoma. Moreover, in recent decades, also in case of pleuro-pericardial metastastes, different but not standardized multimodal therapeutic strategies have been promoted with interesting results [44]. One of the most interesting multimodal approaches is represented by Hyperthermic Intrathoracic Chemotherapy (HITHOC).…”

Section: M-factormentioning

confidence: 99%

TNM Staging System in Thymoma: A Critical Appraisal?

Ambrogi,

Aprile,

Lenzini

et al. 2024

JCM

View full text Add to dashboard Cite

Thymomas are rare tumors of the anterior mediastinum with peculiar clinical and pathological features. They have been deeply analyzed by pioneer authors, who strictly linked their name to the main pathological and staging classifications. Before the latest edition of the WHO classification of thymic epithelial tumors, the history of thymoma pathological classification inherited the name of the pathologists who systematically addressed the issue, from Levine-Rosai to Muller-Hermelink. Similarly, the thymoma staging system is intimately related to the name of two surgeons, Masaoka and Koga, who historically dealt with this disease. More recently, the traditional tumor-nodes-metastasis (TNM) system has been developed for the staging of this condition, in a rational attempt to put thymomas in conformity with the other solid tumors. The efforts of the International Thymic Malignancies Interest Group (ITMIG) and the Thymic Domain of the Staging and Prognostic Factors Committee (TD-SPFC) of the International Association for the Study of Lung Cancer (IASLC) resulted in the TNM classification of thymic tumors, which have been included in the eighth edition of the American Joint Committee on Cancer’s (AJCC) Cancer Staging Manual. Herein, we report a narrative review of the evolution of the thymic epithelial tumors (TET) staging system and present a critical appraisal of the actual TNM classification compared with the historical Masaoka-Koga classification, with special focus on the proposal for the ninth edition of the TNM, expected in 2024.

show abstract

Application of Artificial Intelligence in the Diagnosis, Treatment, and Prognostic Evaluation of Mediastinal Malignant Tumors

Cited by 10 publications

References 123 publications

A special delivery by a fork: Where does artificial intelligence come from?

A special delivery by a fork: Where does artificial intelligence come from?

Effects of Interobserver Segmentation Variability and Intensity Discretization on MRI-Based Radiomic Feature Reproducibility of Lipoma and Atypical Lipomatous Tumor

TNM Staging System in Thymoma: A Critical Appraisal?

Contact Info

Product

Resources

About