|
| Reference | Focus | Differences |
|
| [5] | (i) Diagnosing COVID-19 and predicting severity and mortality risks | (i) The review is based only on clinical and laboratory data. |
| [6] | (i) Diagnosis and prognosis of COVID-19 from prediction models. | (i) The review focuses more on preprints
(ii) It covers all types of models including risk prediction, diagnosis of severity, and diagnosis from images |
| [7] | (i) Diagnosing hepatitis | (i) It covers only clinical tests. |
| [8] | (i) Detecting pneumonia | (i) It is based only on signs and symptoms
(ii) Performance measures are not covered |
| [9] | (i) Diagnosing tuberculosis | (i) It covers diverse AI approaches using clinical signs and symptoms and radiological images. |
| [10] | (i) Diagnosing pulmonary tuberculosis | (i) It covers AI methods based on chest X-ray images. |
| [11] | (i) Diagnosing tuberculous meningitis | (i) It is based only on clinical and laboratory data. |
| [12] | (i) Predicting phenotypic characteristics of influenza virus | (i) It is based on genomic or proteomic input. |
| [13] | (i) Diagnosing HIV, HCV, and chlamydia | (i) It implements different digital technology but does not include any kind of AI technique. |
| [14] | (i) Diagnosing COVID-19, hepatitis, sepsis, malaria, Lyme disease, and tuberculosis | (i) It covers data coming from EMR. |
| [15] | (i) Automatic diagnosis of several infections such as sepsis, general infections, and Clostridium difficile infection through ML and expert system | (i) It covers papers based on physiological data. |
| [16] | (i) Diagnosing infectious and noninfectious diseases through ML | (i) It explains in detail all reviewed ML algorithms but does not mention datasets or performance measures. |
| Our review | (i) ML diagnosis of all available human infectious disease papers | (i) It covers different kinds of ML techniques, several types of datasets, and performance measures. |
|
