1
|
Koseoglu FD, Keklik Karadag F, Bulbul H, Alici EU, Ozyilmaz B, Ozdemir TR. JAKCalc: A machine-learning approach to rationalized JAK2 testing in patients with elevated hemoglobin levels. Medicine (Baltimore) 2024; 103:e37751. [PMID: 38579024 PMCID: PMC10994541 DOI: 10.1097/md.0000000000037751] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 12/18/2023] [Accepted: 03/08/2024] [Indexed: 04/07/2024] Open
Abstract
The demand for Janus Kinase-2 (JAK2) testing has been disproportionate to the low yield of positive results, which highlights the need for more discerning test strategies. The aim of this study is to introduce an artificial intelligence application as a more rational approach for testing JAK2 mutations in cases of erythrocytosis. Test results were sourced from samples sent to a tertiary hospital's genetic laboratory between 2017 and 2023, meeting 2016 World Health Organization criteria for JAK2V617F mutation testing. The JAK2 Somatic Mutation Screening Kit was used for genetic testing. Machine learning models were trained and tested using Python programming language. Out of 458 cases, JAK2V617F mutation was identified in 13.3%. There were significant differences in complete blood count parameters between mutation carriers and non-carriers. Various models were trained with data, with the random forest (RF) model demonstrating superior precision, recall, F1-score, accuracy, and area under the receiver operating characteristic, all reaching 100%. Gradient boosting (GB) model also showed high scores. When compared with existing algorithms, the RF and GB models displayed superior performance. The RF and GB models outperformed other methods in accurately identifying and classifying erythrocytosis cases, offering potential reductions in unnecessary testing and costs.
Collapse
Affiliation(s)
- Fatos Dilan Koseoglu
- Department of Internal Medicine Division of Hematology, Izmir Bakircay University Faculty of Medicine, Cigli Hospital, İzmir, Turkey
| | | | - Hale Bulbul
- Department of Hematology, Tepecik Training and Research Hospital, İzmir, Turkey
| | | | - Berk Ozyilmaz
- Department of Medical Genetics, Tepecik Training and Research Hospital, İzmir, Turkey
| | - Taha Resid Ozdemir
- Department of Medical Genetics, Tepecik Training and Research Hospital, İzmir, Turkey
| |
Collapse
|
2
|
Koseoglu FD, Alıcı IO, Er O. Machine learning approaches in the interpretation of endobronchial ultrasound images: a comparative analysis. Surg Endosc 2023; 37:9339-9346. [PMID: 37903885 DOI: 10.1007/s00464-023-10488-x] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/07/2023] [Accepted: 09/23/2023] [Indexed: 11/01/2023]
Abstract
BACKGROUND This study explores the application of machine learning (ML) in analyzing endobronchial ultrasound (EBUS) images for the detection of lymph node (LN) malignancy, aiming to augment diagnostic accuracy and efficiency. We investigated whether ML could outperform conventional classification systems in identifying malignant involvement of LNs, based on eight established sonographic features. METHODS Retrospective data from two tertiary care hospital bronchoscopy units were utilized, encompassing healthcare reports of patients who had undergone EBUS between January 2017 and March 2023. The ML model was trained and tested using MATLAB, with 80% of the data allocated for training/validation, and 20% for testing. Performance was evaluated based on validation and testing accuracy, and receiver operating characteristic curves with comparing trained models and existing classification rules. RESULTS The study analyzed 992 LNs, with 42.3% malignancy prevalence. Malignant LNs showed characteristic features such as larger size and distinct margins. The fine tuned models achieved testing accuracies of 95.9% and 96.4% for fine Gaussian SVM and KNN, respectively. Corresponding AUROC's were 0.955 and 0.963, outperforming other similar studies and conventional analyses. CONCLUSION Fine tuned ML applications like SVM and KNN, can significantly enhance the analysis of EBUS images, improving diagnostic accuracy.
Collapse
Affiliation(s)
- Fatos Dilan Koseoglu
- Department of Internal Medicine Division of Hematology, Cigli Hospital, Izmir Bakircay University, Yeni District, 8780/1. Str. No:18, 35620, Ciğli, İzmir, Turkey.
| | - Ibrahim Onur Alıcı
- Department of Pulmonary Medicine, Faculty of Medicine, Izmir Bakircay University, İzmir, Turkey
| | - Orhan Er
- Department of Computer Engineering, Faculty of Architecture and Engineering, Izmir Bakircay University, İzmir, Turkey
| |
Collapse
|
3
|
Akinci B, Koseoglu FD, Onay H, Yavuz S, Altay C, Simsir IY, Ozisik S, Demir L, Korkut M, Yilmaz N, Ozen S, Akinci G, Atik T, Calan M, Secil M, Comlekci A, Demir T. Acquired partial lipodystrophy is associated with increased risk for developing metabolic abnormalities. Metabolism 2015; 64:1086-95. [PMID: 26139569 DOI: 10.1016/j.metabol.2015.06.004] [Citation(s) in RCA: 20] [Impact Index Per Article: 2.2] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 02/02/2015] [Revised: 06/01/2015] [Accepted: 06/04/2015] [Indexed: 02/05/2023]
Abstract
OBJECTIVE Acquired partial lipodystrophy (APL) is a rare disorder characterized by progressive selective fat loss. In previous studies, metabolic abnormalities were reported to be relatively rare in APL, whilst they were quite common in other types of lipodystrophy syndromes. METHODS In this nationwide cohort study, we evaluated 21 Turkish patients with APL who were enrolled in a prospective follow-up protocol. Subjects were investigated for metabolic abnormalities. Fat distribution was assessed by whole body MRI. Hepatic steatosis was evaluated by ultrasound, MRI and MR spectroscopy. Patients with diabetes underwent a mix meal stimulated C-peptide/insulin test to investigate pancreatic beta cell functions. Leptin and adiponectin levels were measured. RESULTS Fifteen individuals (71.4%) had at least one metabolic abnormality. Six patients (28.6%) had diabetes, 12 (57.1%) hypertrigylceridemia, 10 (47.6%) low HDL cholesterol, and 11 (52.4%) hepatic steatosis. Steatohepatitis was further confirmed in 2 patients with liver biopsy. Anti-GAD was negative in all APL patients with diabetes. APL patients with diabetes had lower leptin and adiponectin levels compared to patients with type 2 diabetes and healthy controls. However, contrary to what we observed in patients with congenital generalized lipodystrophy (CGL), we did not detect consistently very low leptin levels in APL patients. The mix meal test suggested that APL patients with diabetes had a significant amount of functional pancreatic beta cells, and their diabetes was apparently associated with insulin resistance. CONCLUSIONS Our results show that APL is associated with increased risk for developing metabolic abnormalities. We suggest that close long-term follow-up is required to identify and manage metabolic abnormalities in APL.
Collapse
Affiliation(s)
- Baris Akinci
- Dokuz Eylul University, Division of Endocrinology, Izmir, Turkey.
| | | | - Huseyin Onay
- Ege University, Department of Medical Genetics, Izmir, Turkey
| | - Sevgi Yavuz
- Kanuni Sultan Suleyman Training Hospital, Department of Dermatology, Istanbul, Turkey
| | - Canan Altay
- Dokuz Eylul University, Department of Radiology, Izmir, Turkey
| | | | - Secil Ozisik
- Dokuz Eylul University, Division of Endocrinology, Izmir, Turkey
| | - Leyla Demir
- Ataturk Training Hospital, Department of Biochemistry, Izmir, Turkey
| | - Meltem Korkut
- Yeditepe University, Division of Pediatric Gastroenterology, Istanbul, Turkey
| | - Nusret Yilmaz
- Akdeniz University, Division of Endocrinology, Antalya, Turkey
| | - Samim Ozen
- Ege University, Department of Medical Genetics, Izmir, Turkey; Ege University, Division of Pediatric Endocrinology, Izmir, Turkey
| | - Gulcin Akinci
- Dr.Behcet Uz Children's Hospital, Division of Pediatric Neurology, Izmir, Turkey
| | - Tahir Atik
- Ege University, Department of Medical Genetics, Izmir, Turkey
| | - Mehmet Calan
- Dokuz Eylul University, Division of Endocrinology, Izmir, Turkey
| | - Mustafa Secil
- Dokuz Eylul University, Department of Radiology, Izmir, Turkey
| | | | - Tevfik Demir
- Dokuz Eylul University, Division of Endocrinology, Izmir, Turkey
| |
Collapse
|
4
|
Koseoglu FD, Arslan C. Bisphosphonate therapy in metastatic carcinoma patients with chronic renal failure: are bisphosphonates an enemy or crony? Support Care Cancer 2015; 23:1489-91. [PMID: 25763753 DOI: 10.1007/s00520-015-2672-2] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/03/2014] [Accepted: 02/22/2015] [Indexed: 11/25/2022]
Affiliation(s)
- Fatos Dilan Koseoglu
- Department of Internal Medicine, Tepecik Education and Research Hospital, Gaziler Caddesi No: 468, 35170, Konak, Izmir, Turkey,
| | | |
Collapse
|
5
|
Arslan C, Koseoglu FD. Modified docetaxel, cisplatin, and 5-fluorouracil combination regimen in advanced gastric cancer: Toxicity and efficacy results. J Clin Oncol 2014. [DOI: 10.1200/jco.2014.32.15_suppl.e15036] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.1] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/20/2022] Open
Affiliation(s)
- Cagatay Arslan
- Izmir University MedicalPark Hospital Department of Medical Oncology, Izmir, Turkey
| | - Fatos Dilan Koseoglu
- Izmir Tepecik Research and Training Hospital, Department of Internal Medicine, Izmir, Turkey
| |
Collapse
|