Research for Early Cancer Detection & Accurate Identification

Early cancer detection and accurate cancer type and subtype identification are imperative for better patient outcomes and should increase the life span and improve the quality of life for some of these affected patients. In order to detect lung cancer, standard procedures such as microscopic examination of tissue slides and pathology image analysis have been in practice for years. However, these methods are not only time-consuming but also provide a lot of variabilities in result interpretation (Wang et al., 2019). To make this process more precise and reduce variations, AI should be applied.

ML in genomic oncology. One way that ML can be utilized in lung cancer diagnosis and treatment is the genomic classification of NSCLC, where microarrays are used to train ML algorithms to identify different lung cancer mutations. This is a vital step since personalized medicine can be of great help in this case, and new treatments could be applied based on the existing mutations (Xu et al., 2019a). This can be especially beneficial for patients with ADC since 20% of lung ADCs are caused by mutations such as those of the epidermal growth factor receptor (EGFR) gene or rearrangements of the anaplastic lymphoma kinase (ALK) gene which are already being used in personalized medicine as biomarkers for targeted treatments (Cagle, Raparia, & Portier, 2016; Mascaux, Tomasini, Greillier, & Barlesi, 2017).

In addition to EGFR and ALK, mutations such as KRAS, BRAF, HER2, MET or translocation of ROS1 and RET genes are also being identified in lung ADC as potential biomarkers for targeted treatments, and are being used as targets in various clinical trials (Cagle et al.

Get quality help now

WriterBelle

Verified

Proficient in: Cancer

4.7 (657)

“ Really polite, and a great writer! Task done as described and better, responded to all my questions promptly too! ”

+84 relevant experts are online

Hire writer

, 2016). Being able to use ML to detect these biomarkers is valuable for lung cancer patients who will be not only able to receive the greatest benefits from targeted treatments but will also receive treatment resulting in lower toxicity (Mascaux et al., 2017). In regard to SCC patients, targeted therapies are unfortunately still not available, and treatments used for ADC patients are deemed ineffective; however, ML is being used for detecting mutations that could possibly be used as biomarkers for targeted treatments.

ML in treatment personalization. Besides being used in lung cancer diagnosis, ML is being implemented in determining personalized treatment plans in lung cancer as well as following the prognosis. When deciding on a personalized treatment plan and patient survival, it is vital to distinguish between the two main sub-types of NSCLCs, SCC and ADC, since patients with different NSCLC sub-types require distinctive treatment approaches and have different survival outcomes (Zappa & Mousa, 2016). In regard to the treatment plan, this characterization is vital since some chemotherapy treatments have been reported to be more successful in treating SCC while other treatments are more successful in treating ADC (Huang et al., 2016; Shoshan-Barmatz et al., 2017).

In addition, ML is used to extract quantitative features, such as cell size and shape, scattering of pixel intensity in the cells and nuclei, and texture of the cells and nuclei, from histological slides, which, in turn, will help with distinguishing between cancerous tissue and surrounding healthy tissue (Yu et al., 2016). Once extracted, these aforementioned quantitative image features in combination with the patient survival indices can be advantageous in determining patient survival outcomes. When predicting survival outcomes in patients with ADC, texture of the nuclei, Zernike shape decomposition of the nuclei, and Zernike shape decomposition of the cytoplasm are of a great interest; while predicting survival outcomes in patients with SCC, Zernike shape in the tumor nuclei and cytoplasm are being considered (Yu et al., 2016). Additionally, patient features such as genetic, biochemical, physiological, behavioral, as well as environmental exposure are all taken into consideration when AI and ML are used in predicting the survival outcome of lung cancer patients (Schork, 2019).

DL in digital imaging and treatment response prediction. DL is a sub-discipline of ML, and one area of pulmonary oncology where application of DL-based algorithms has great potential relating to the prediction of treatment response for NSCLC patients who were undergoing chemotherapy and radiation treatment. Predictions can be achieved by analyzing time series computed tomography (CT) images of this group of patients (Xu et al., 2019b). In this case, a series of lung CT images are taken starting with the baseline images or images taken prior to the patient starting the treatment, post treatment images or images taken after each treatment, and follow-up images or images taken after the patient has completed the treatment, are compared and monitored for any changes that may happen during and post treatment.

This can be of great help for pulmonary oncologists and lung pathologists, who will be able to measure tumor diameter from CT images, and who will be able, if necessary, to modify treatments for their patients to make them more personalized, which in turn can improve the outcome of those patients (Bi et al., 2019). In addition, any errors that could potentially occur due to the manually comparing of CT images will be minimized while efficiency will be increased.

DL in lung cancer diagnosis. DL can also be used in cancer diagnosis. In this case, DL is applied to whole slide imaging (WSI), where high resolution digital images of lung cancer slides were generated, patterns are observed, and if present, cancer could be detected (Mukhopadhyay, 2018). This digital WSI was tested and compared to microscopy currently used in cancer diagnosis, and it has been found that the tissue analysis results were comparable (Mukhopadhyay, 2018). DL use in this case can reduce potential errors, and hopefully it will gain bigger popularity in the future.

NLP in evaluation of disease status and patient response to therapies. NLP is a subcategory of AI which can be used for analyzing unstructured data in pulmonary oncology. Once unstructured data is extracted by computers and machines from clinical notes, clinical laboratory reports, pathology reports, surgical notes, and other clinical sources, NLP is used to convert this data to a structured one which can be then analyzed by ML (Jiang et al., 2017). In pulmonary oncology, NLP is primarily being used in assessing lung cancer progression and patient response to given therapies based on the radiology reports (Kehl et al., 2019). In this case, NLP extracts clinically relevant oncologic data such as cancer status (regression, stable, progress), magnitude of change (mild, moderate, marked), and the significance of change (uncertain, possible, probable) from radiology reports, and then encodes this data in a structured format which would be easier for ML to analyze (Cheng, Zheng, Savova, & Erickson, 2010).

NLP in prediction of patient survival rates and patient response to treatment. In addition, NLP is used for extracting unstructured data from clinical encounters, treatments, prognosis, and follow-up notes of patients undergoing chemoradiotherapy, which can help monitor the patient’s response to treatment as well as predict patient survival rates (Zheng et al., 2018). In this case, Information and Data Extraction using Adaptive Learning (IDEAL-X) system could be used, and treatment site, chemotherapy information, treatment time frame, radiation therapy dose, and toxicities could be automatically extracted from multiple pages within one second (Zheng et al., 2018).

IDEAL-X system has been tested and it has been reported that the overall precision in regard to extraction of patient characteristics and tumor control is over 93% while overall precision in relation to the extraction of the toxicities is 95.7% (Zheng et al., 2018). Advantages of using IDEAL-X system for data extraction are that it is efficient and extremely accurate, and that accuracy can be even more improved by continuous training (Zheng et al., 2018). Despite being a promising data extraction tool in pulmonary oncology, NLP is not used as often as ML and DL since NLP use is still a new technology. However, with the constant advancements in technology, NLP could provide more significant patient-related data for use in personalized medicine. All this evidence demonstrates the potential that ML, DL, and NLP could have in personalized medicine, and especially in pulmonary oncology.

Secondary/Tertiary Findings

DL in radiotherapy dose customization. Aside from the previously mentioned possibilities for AI applications in pulmonary oncology, AI and particularly DL could possibly be used in radiotherapy dose customization for each lung cancer patients who will undergo radiation therapy (Lou et al., 2019). This method has been verified in a clinical environment, and the obtained results have shown that DL could be successfully used in customizing radiotherapy dose for each patient based on their lung images and clinical variables from electronic health records (EHR), and that different patients express different levels of sensitivity to radiotherapy (Lou et al., 2019). If more clinical testing is done and the results are consistent, this method could be extremely important for patients since not only toxicity could be reduced for them, but also patient outcomes could be improved.

DL in biomarker detection. Another possible approach of using AI in lung oncology and precision medicine that is being explored in the clinical environment refers to using DL in immuno-histochemistry (IHC) for detection of programmed death-ligand 1 (PD-L1) biomarker (Mascaux et al., 2017; Sha et al., 2019). In this case, DL could be a valuable tool in predicting PD-L1 status from hematoxylin and eosin (H and E) whole-slide images (WSIs) of lung cancer patients’ tumor samples. PD-L1 biomarker enables identification of those lung cancer patients who would benefit the most from checkpoint inhibitor immune checkpoint inhibitor (ICI) therapy, and testing for this biomarker is suggested for those patients that have metastatic NSCLC and that have been tested for EGFR, ALK, and ROS1 mutations with negative or undetermined results (Mascaux et al., 2017).

Exploration of this technique could benefit in cases when not enough tissue is available for further testing or when there is a lack of reagents required for IHC staining (Sha et al., 2019). Study results were promising and indicate that expression of PD-L1 biomarker is associated with the morphological characteristics of the tumor microenvironment. Even though obtained results were highly accurate, further testing might be necessary to verify these findings.